{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "80979729-8f2a-4877-a977-2ce88ec57ae9",
   "metadata": {},
   "source": [
    "# Classification"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "478f73e2-30eb-4990-8b4b-b56de4bc17b9",
   "metadata": {},
   "source": [
    "<img src=\"https://blog.otto-office.com/wp-content/uploads/2018/12/kuendigung.jpg\" width=850>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "562fe7f0-9b37-49bc-849b-fecd9336252f",
   "metadata": {},
   "source": [
    "### Dataset Download  \n",
    "You can download the CSV file here:  \n",
    "[https://www.kaggle.com/competitions/playground-series-s3e3/data)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bcfd3781-cd13-452d-9c14-a6e5009f7760",
   "metadata": {},
   "source": [
    "### Introduction\n",
    "Employee attrition is an important business problem, as high turnover can negatively impact productivity and costs.\n",
    "In this project, a binary classification model is built to predict the probability of employee attrition using tabular employee data.\n",
    "The model is evaluated using ROC AUC, which measures how well the model separates employees who are likely to leave from those who are not.\n",
    "The goal is to develop a reliable and well-validated model following machine learning best practices."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e1f54cb-0660-46b5-baaf-4d010838a1c5",
   "metadata": {},
   "source": [
    "### Import Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "49bfb84a-de8e-4e02-a268-0184108d29f9",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split,StratifiedKFold\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix, classification_report\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from catboost import CatBoostClassifier\n",
    "from lightgbm import LGBMClassifier\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.metrics import roc_auc_score\n",
    "from sklearn.metrics import roc_curve,auc\n",
    "import joblib\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "pd.set_option('display.max_columns',100)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1642c795-31ff-43b0-ad7b-dc1c2bddd837",
   "metadata": {},
   "source": [
    "### Load Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "274379ab-4c1b-47b8-8970-1e73a756ec8a",
   "metadata": {},
   "outputs": [],
   "source": [
    "df1=pd.read_csv('train.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b3756818-7e57-4be3-b4db-1f800e3c5261",
   "metadata": {},
   "outputs": [],
   "source": [
    "df2=pd.read_csv('test.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19cb289e-990d-4816-b928-8bfc11651fd0",
   "metadata": {},
   "source": [
    "### EDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "82f8d7dd-34ab-49a9-abaa-f1256dbe30e4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>Age</th>\n",
       "      <th>BusinessTravel</th>\n",
       "      <th>DailyRate</th>\n",
       "      <th>Department</th>\n",
       "      <th>DistanceFromHome</th>\n",
       "      <th>Education</th>\n",
       "      <th>EducationField</th>\n",
       "      <th>EmployeeCount</th>\n",
       "      <th>EnvironmentSatisfaction</th>\n",
       "      <th>Gender</th>\n",
       "      <th>HourlyRate</th>\n",
       "      <th>JobInvolvement</th>\n",
       "      <th>JobLevel</th>\n",
       "      <th>JobRole</th>\n",
       "      <th>JobSatisfaction</th>\n",
       "      <th>MaritalStatus</th>\n",
       "      <th>MonthlyIncome</th>\n",
       "      <th>MonthlyRate</th>\n",
       "      <th>NumCompaniesWorked</th>\n",
       "      <th>Over18</th>\n",
       "      <th>OverTime</th>\n",
       "      <th>PercentSalaryHike</th>\n",
       "      <th>PerformanceRating</th>\n",
       "      <th>RelationshipSatisfaction</th>\n",
       "      <th>StandardHours</th>\n",
       "      <th>StockOptionLevel</th>\n",
       "      <th>TotalWorkingYears</th>\n",
       "      <th>TrainingTimesLastYear</th>\n",
       "      <th>WorkLifeBalance</th>\n",
       "      <th>YearsAtCompany</th>\n",
       "      <th>YearsInCurrentRole</th>\n",
       "      <th>YearsSinceLastPromotion</th>\n",
       "      <th>YearsWithCurrManager</th>\n",
       "      <th>Attrition</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>36</td>\n",
       "      <td>Travel_Frequently</td>\n",
       "      <td>599</td>\n",
       "      <td>Research &amp; Development</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>Medical</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>Male</td>\n",
       "      <td>42</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Laboratory Technician</td>\n",
       "      <td>4</td>\n",
       "      <td>Married</td>\n",
       "      <td>2596</td>\n",
       "      <td>5099</td>\n",
       "      <td>1</td>\n",
       "      <td>Y</td>\n",
       "      <td>Yes</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>921</td>\n",
       "      <td>Sales</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>Other</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "      <td>46</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Sales Representative</td>\n",
       "      <td>1</td>\n",
       "      <td>Married</td>\n",
       "      <td>2899</td>\n",
       "      <td>10778</td>\n",
       "      <td>1</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>17</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>32</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>718</td>\n",
       "      <td>Sales</td>\n",
       "      <td>26</td>\n",
       "      <td>3</td>\n",
       "      <td>Marketing</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Male</td>\n",
       "      <td>80</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>Sales Executive</td>\n",
       "      <td>4</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>4627</td>\n",
       "      <td>16495</td>\n",
       "      <td>0</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>17</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>80</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>38</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>1488</td>\n",
       "      <td>Research &amp; Development</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>Medical</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Female</td>\n",
       "      <td>40</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>Healthcare Representative</td>\n",
       "      <td>1</td>\n",
       "      <td>Married</td>\n",
       "      <td>5347</td>\n",
       "      <td>13384</td>\n",
       "      <td>3</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>50</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>1017</td>\n",
       "      <td>Research &amp; Development</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>Medical</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>Female</td>\n",
       "      <td>37</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>Manager</td>\n",
       "      <td>1</td>\n",
       "      <td>Single</td>\n",
       "      <td>19033</td>\n",
       "      <td>19805</td>\n",
       "      <td>1</td>\n",
       "      <td>Y</td>\n",
       "      <td>Yes</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>31</td>\n",
       "      <td>14</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  Age     BusinessTravel  DailyRate              Department  \\\n",
       "0   0   36  Travel_Frequently        599  Research & Development   \n",
       "1   1   35      Travel_Rarely        921                   Sales   \n",
       "2   2   32      Travel_Rarely        718                   Sales   \n",
       "3   3   38      Travel_Rarely       1488  Research & Development   \n",
       "4   4   50      Travel_Rarely       1017  Research & Development   \n",
       "\n",
       "   DistanceFromHome  Education EducationField  EmployeeCount  \\\n",
       "0                24          3        Medical              1   \n",
       "1                 8          3          Other              1   \n",
       "2                26          3      Marketing              1   \n",
       "3                 2          3        Medical              1   \n",
       "4                 5          4        Medical              1   \n",
       "\n",
       "   EnvironmentSatisfaction  Gender  HourlyRate  JobInvolvement  JobLevel  \\\n",
       "0                        4    Male          42               3         1   \n",
       "1                        1    Male          46               3         1   \n",
       "2                        3    Male          80               3         2   \n",
       "3                        3  Female          40               3         2   \n",
       "4                        2  Female          37               3         5   \n",
       "\n",
       "                     JobRole  JobSatisfaction MaritalStatus  MonthlyIncome  \\\n",
       "0      Laboratory Technician                4       Married           2596   \n",
       "1       Sales Representative                1       Married           2899   \n",
       "2            Sales Executive                4      Divorced           4627   \n",
       "3  Healthcare Representative                1       Married           5347   \n",
       "4                    Manager                1        Single          19033   \n",
       "\n",
       "   MonthlyRate  NumCompaniesWorked Over18 OverTime  PercentSalaryHike  \\\n",
       "0         5099                   1      Y      Yes                 13   \n",
       "1        10778                   1      Y       No                 17   \n",
       "2        16495                   0      Y       No                 17   \n",
       "3        13384                   3      Y       No                 14   \n",
       "4        19805                   1      Y      Yes                 13   \n",
       "\n",
       "   PerformanceRating  RelationshipSatisfaction  StandardHours  \\\n",
       "0                  3                         2             80   \n",
       "1                  3                         4             80   \n",
       "2                  3                         4             80   \n",
       "3                  3                         3             80   \n",
       "4                  3                         3             80   \n",
       "\n",
       "   StockOptionLevel  TotalWorkingYears  TrainingTimesLastYear  \\\n",
       "0                 1                 10                      2   \n",
       "1                 1                  4                      3   \n",
       "2                 2                  4                      3   \n",
       "3                 0                 15                      1   \n",
       "4                 0                 31                      0   \n",
       "\n",
       "   WorkLifeBalance  YearsAtCompany  YearsInCurrentRole  \\\n",
       "0                3              10                   0   \n",
       "1                3               4                   2   \n",
       "2                3               3                   2   \n",
       "3                1               6                   0   \n",
       "4                3              31                  14   \n",
       "\n",
       "   YearsSinceLastPromotion  YearsWithCurrManager  Attrition  \n",
       "0                        7                     8          0  \n",
       "1                        0                     3          0  \n",
       "2                        1                     2          0  \n",
       "3                        0                     2          0  \n",
       "4                        4                    10          1  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "43f40647-8d39-48d3-920d-9335bd88d57a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>Age</th>\n",
       "      <th>BusinessTravel</th>\n",
       "      <th>DailyRate</th>\n",
       "      <th>Department</th>\n",
       "      <th>DistanceFromHome</th>\n",
       "      <th>Education</th>\n",
       "      <th>EducationField</th>\n",
       "      <th>EmployeeCount</th>\n",
       "      <th>EnvironmentSatisfaction</th>\n",
       "      <th>Gender</th>\n",
       "      <th>HourlyRate</th>\n",
       "      <th>JobInvolvement</th>\n",
       "      <th>JobLevel</th>\n",
       "      <th>JobRole</th>\n",
       "      <th>JobSatisfaction</th>\n",
       "      <th>MaritalStatus</th>\n",
       "      <th>MonthlyIncome</th>\n",
       "      <th>MonthlyRate</th>\n",
       "      <th>NumCompaniesWorked</th>\n",
       "      <th>Over18</th>\n",
       "      <th>OverTime</th>\n",
       "      <th>PercentSalaryHike</th>\n",
       "      <th>PerformanceRating</th>\n",
       "      <th>RelationshipSatisfaction</th>\n",
       "      <th>StandardHours</th>\n",
       "      <th>StockOptionLevel</th>\n",
       "      <th>TotalWorkingYears</th>\n",
       "      <th>TrainingTimesLastYear</th>\n",
       "      <th>WorkLifeBalance</th>\n",
       "      <th>YearsAtCompany</th>\n",
       "      <th>YearsInCurrentRole</th>\n",
       "      <th>YearsSinceLastPromotion</th>\n",
       "      <th>YearsWithCurrManager</th>\n",
       "      <th>Attrition</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1672</th>\n",
       "      <td>1672</td>\n",
       "      <td>30</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>945</td>\n",
       "      <td>Sales</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Life Sciences</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>Female</td>\n",
       "      <td>73</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>Sales Executive</td>\n",
       "      <td>3</td>\n",
       "      <td>Single</td>\n",
       "      <td>8722</td>\n",
       "      <td>14255</td>\n",
       "      <td>1</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>19</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1673</th>\n",
       "      <td>1673</td>\n",
       "      <td>32</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>1303</td>\n",
       "      <td>Research &amp; Development</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>Life Sciences</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "      <td>48</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Research Scientist</td>\n",
       "      <td>2</td>\n",
       "      <td>Married</td>\n",
       "      <td>3544</td>\n",
       "      <td>15972</td>\n",
       "      <td>4</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>19</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1674</th>\n",
       "      <td>1674</td>\n",
       "      <td>29</td>\n",
       "      <td>Travel_Frequently</td>\n",
       "      <td>1184</td>\n",
       "      <td>Human Resources</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>Human Resources</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>Male</td>\n",
       "      <td>36</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>Human Resources</td>\n",
       "      <td>1</td>\n",
       "      <td>Married</td>\n",
       "      <td>2804</td>\n",
       "      <td>15322</td>\n",
       "      <td>1</td>\n",
       "      <td>Y</td>\n",
       "      <td>Yes</td>\n",
       "      <td>11</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1675</th>\n",
       "      <td>1675</td>\n",
       "      <td>36</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>441</td>\n",
       "      <td>Sales</td>\n",
       "      <td>9</td>\n",
       "      <td>2</td>\n",
       "      <td>Marketing</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>Male</td>\n",
       "      <td>48</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>Sales Executive</td>\n",
       "      <td>3</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>5406</td>\n",
       "      <td>4051</td>\n",
       "      <td>1</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>21</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>80</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1676</th>\n",
       "      <td>1676</td>\n",
       "      <td>36</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>1141</td>\n",
       "      <td>Research &amp; Development</td>\n",
       "      <td>20</td>\n",
       "      <td>3</td>\n",
       "      <td>Life Sciences</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Female</td>\n",
       "      <td>35</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Laboratory Technician</td>\n",
       "      <td>3</td>\n",
       "      <td>Single</td>\n",
       "      <td>2593</td>\n",
       "      <td>17381</td>\n",
       "      <td>0</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>19</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        id  Age     BusinessTravel  DailyRate              Department  \\\n",
       "1672  1672   30      Travel_Rarely        945                   Sales   \n",
       "1673  1673   32      Travel_Rarely       1303  Research & Development   \n",
       "1674  1674   29  Travel_Frequently       1184         Human Resources   \n",
       "1675  1675   36      Travel_Rarely        441                   Sales   \n",
       "1676  1676   36      Travel_Rarely       1141  Research & Development   \n",
       "\n",
       "      DistanceFromHome  Education   EducationField  EmployeeCount  \\\n",
       "1672                 1          3    Life Sciences              1   \n",
       "1673                 2          3    Life Sciences              1   \n",
       "1674                24          3  Human Resources              1   \n",
       "1675                 9          2        Marketing              1   \n",
       "1676                20          3    Life Sciences              1   \n",
       "\n",
       "      EnvironmentSatisfaction  Gender  HourlyRate  JobInvolvement  JobLevel  \\\n",
       "1672                        4  Female          73               3         3   \n",
       "1673                        1    Male          48               3         1   \n",
       "1674                        2    Male          36               2         1   \n",
       "1675                        2    Male          48               4         2   \n",
       "1676                        3  Female          35               3         1   \n",
       "\n",
       "                    JobRole  JobSatisfaction MaritalStatus  MonthlyIncome  \\\n",
       "1672        Sales Executive                3        Single           8722   \n",
       "1673     Research Scientist                2       Married           3544   \n",
       "1674        Human Resources                1       Married           2804   \n",
       "1675        Sales Executive                3      Divorced           5406   \n",
       "1676  Laboratory Technician                3        Single           2593   \n",
       "\n",
       "      MonthlyRate  NumCompaniesWorked Over18 OverTime  PercentSalaryHike  \\\n",
       "1672        14255                   1      Y       No                 19   \n",
       "1673        15972                   4      Y       No                 19   \n",
       "1674        15322                   1      Y      Yes                 11   \n",
       "1675         4051                   1      Y       No                 21   \n",
       "1676        17381                   0      Y       No                 19   \n",
       "\n",
       "      PerformanceRating  RelationshipSatisfaction  StandardHours  \\\n",
       "1672                  3                         2             80   \n",
       "1673                  3                         4             80   \n",
       "1674                  3                         3             80   \n",
       "1675                  4                         3             80   \n",
       "1676                  3                         4             80   \n",
       "\n",
       "      StockOptionLevel  TotalWorkingYears  TrainingTimesLastYear  \\\n",
       "1672                 0                 10                      2   \n",
       "1673                 1                 10                      3   \n",
       "1674                 0                  1                      2   \n",
       "1675                 2                 10                      3   \n",
       "1676                 0                 10                      3   \n",
       "\n",
       "      WorkLifeBalance  YearsAtCompany  YearsInCurrentRole  \\\n",
       "1672                4              10                   0   \n",
       "1673                4               4                   2   \n",
       "1674                3               1                   0   \n",
       "1675                2              10                   3   \n",
       "1676                2               8                   2   \n",
       "\n",
       "      YearsSinceLastPromotion  YearsWithCurrManager  Attrition  \n",
       "1672                        0                     8          0  \n",
       "1673                        1                     3          0  \n",
       "1674                        0                     0          1  \n",
       "1675                        0                     8          0  \n",
       "1676                        7                     3          0  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "983b5ad8-3341-4c15-aa80-3476f3905b34",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>Age</th>\n",
       "      <th>BusinessTravel</th>\n",
       "      <th>DailyRate</th>\n",
       "      <th>Department</th>\n",
       "      <th>DistanceFromHome</th>\n",
       "      <th>Education</th>\n",
       "      <th>EducationField</th>\n",
       "      <th>EmployeeCount</th>\n",
       "      <th>EnvironmentSatisfaction</th>\n",
       "      <th>Gender</th>\n",
       "      <th>HourlyRate</th>\n",
       "      <th>JobInvolvement</th>\n",
       "      <th>JobLevel</th>\n",
       "      <th>JobRole</th>\n",
       "      <th>JobSatisfaction</th>\n",
       "      <th>MaritalStatus</th>\n",
       "      <th>MonthlyIncome</th>\n",
       "      <th>MonthlyRate</th>\n",
       "      <th>NumCompaniesWorked</th>\n",
       "      <th>Over18</th>\n",
       "      <th>OverTime</th>\n",
       "      <th>PercentSalaryHike</th>\n",
       "      <th>PerformanceRating</th>\n",
       "      <th>RelationshipSatisfaction</th>\n",
       "      <th>StandardHours</th>\n",
       "      <th>StockOptionLevel</th>\n",
       "      <th>TotalWorkingYears</th>\n",
       "      <th>TrainingTimesLastYear</th>\n",
       "      <th>WorkLifeBalance</th>\n",
       "      <th>YearsAtCompany</th>\n",
       "      <th>YearsInCurrentRole</th>\n",
       "      <th>YearsSinceLastPromotion</th>\n",
       "      <th>YearsWithCurrManager</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1677</td>\n",
       "      <td>19</td>\n",
       "      <td>Non-Travel</td>\n",
       "      <td>992</td>\n",
       "      <td>Research &amp; Development</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Medical</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>Male</td>\n",
       "      <td>43</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Laboratory Technician</td>\n",
       "      <td>3</td>\n",
       "      <td>Single</td>\n",
       "      <td>2318</td>\n",
       "      <td>17778</td>\n",
       "      <td>1</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>12</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1678</td>\n",
       "      <td>45</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>1136</td>\n",
       "      <td>Sales</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>Marketing</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Male</td>\n",
       "      <td>67</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>Sales Executive</td>\n",
       "      <td>1</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>5486</td>\n",
       "      <td>12421</td>\n",
       "      <td>6</td>\n",
       "      <td>Y</td>\n",
       "      <td>Yes</td>\n",
       "      <td>12</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1679</td>\n",
       "      <td>37</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>155</td>\n",
       "      <td>Research &amp; Development</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>Life Sciences</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>Male</td>\n",
       "      <td>41</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Research Scientist</td>\n",
       "      <td>4</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>2741</td>\n",
       "      <td>23577</td>\n",
       "      <td>4</td>\n",
       "      <td>Y</td>\n",
       "      <td>Yes</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>80</td>\n",
       "      <td>2</td>\n",
       "      <td>13</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1680</td>\n",
       "      <td>32</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>688</td>\n",
       "      <td>Research &amp; Development</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>Life Sciences</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Male</td>\n",
       "      <td>89</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>Healthcare Representative</td>\n",
       "      <td>3</td>\n",
       "      <td>Single</td>\n",
       "      <td>5228</td>\n",
       "      <td>20364</td>\n",
       "      <td>1</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>14</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>14</td>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1681</td>\n",
       "      <td>29</td>\n",
       "      <td>Travel_Frequently</td>\n",
       "      <td>464</td>\n",
       "      <td>Research &amp; Development</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>Life Sciences</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Male</td>\n",
       "      <td>79</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>Laboratory Technician</td>\n",
       "      <td>4</td>\n",
       "      <td>Single</td>\n",
       "      <td>1223</td>\n",
       "      <td>15178</td>\n",
       "      <td>1</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     id  Age     BusinessTravel  DailyRate              Department  \\\n",
       "0  1677   19         Non-Travel        992  Research & Development   \n",
       "1  1678   45      Travel_Rarely       1136                   Sales   \n",
       "2  1679   37      Travel_Rarely        155  Research & Development   \n",
       "3  1680   32      Travel_Rarely        688  Research & Development   \n",
       "4  1681   29  Travel_Frequently        464  Research & Development   \n",
       "\n",
       "   DistanceFromHome  Education EducationField  EmployeeCount  \\\n",
       "0                 1          1        Medical              1   \n",
       "1                 4          4      Marketing              1   \n",
       "2                13          3  Life Sciences              1   \n",
       "3                 1          4  Life Sciences              1   \n",
       "4                 9          1  Life Sciences              1   \n",
       "\n",
       "   EnvironmentSatisfaction Gender  HourlyRate  JobInvolvement  JobLevel  \\\n",
       "0                        4   Male          43               3         1   \n",
       "1                        3   Male          67               3         2   \n",
       "2                        4   Male          41               3         1   \n",
       "3                        3   Male          89               2         2   \n",
       "4                        3   Male          79               3         1   \n",
       "\n",
       "                     JobRole  JobSatisfaction MaritalStatus  MonthlyIncome  \\\n",
       "0      Laboratory Technician                3        Single           2318   \n",
       "1            Sales Executive                1      Divorced           5486   \n",
       "2         Research Scientist                4      Divorced           2741   \n",
       "3  Healthcare Representative                3        Single           5228   \n",
       "4      Laboratory Technician                4        Single           1223   \n",
       "\n",
       "   MonthlyRate  NumCompaniesWorked Over18 OverTime  PercentSalaryHike  \\\n",
       "0        17778                   1      Y       No                 12   \n",
       "1        12421                   6      Y      Yes                 12   \n",
       "2        23577                   4      Y      Yes                 13   \n",
       "3        20364                   1      Y       No                 13   \n",
       "4        15178                   1      Y       No                 14   \n",
       "\n",
       "   PerformanceRating  RelationshipSatisfaction  StandardHours  \\\n",
       "0                  3                         4             80   \n",
       "1                  3                         3             80   \n",
       "2                  3                         2             80   \n",
       "3                  3                         3             80   \n",
       "4                  3                         1             80   \n",
       "\n",
       "   StockOptionLevel  TotalWorkingYears  TrainingTimesLastYear  \\\n",
       "0                 0                  1                      2   \n",
       "1                 1                  7                      3   \n",
       "2                 2                 13                      2   \n",
       "3                 0                 14                      2   \n",
       "4                 0                  1                      5   \n",
       "\n",
       "   WorkLifeBalance  YearsAtCompany  YearsInCurrentRole  \\\n",
       "0                2               1                   0   \n",
       "1                3               2                   2   \n",
       "2                2               7                   7   \n",
       "3                2              14                  10   \n",
       "4                3               1                   0   \n",
       "\n",
       "   YearsSinceLastPromotion  YearsWithCurrManager  \n",
       "0                        0                     0  \n",
       "1                        2                     2  \n",
       "2                        1                     7  \n",
       "3                       11                     8  \n",
       "4                        0                     0  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "61dddcdc-221d-409c-bf69-1b2d99833d49",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>Age</th>\n",
       "      <th>BusinessTravel</th>\n",
       "      <th>DailyRate</th>\n",
       "      <th>Department</th>\n",
       "      <th>DistanceFromHome</th>\n",
       "      <th>Education</th>\n",
       "      <th>EducationField</th>\n",
       "      <th>EmployeeCount</th>\n",
       "      <th>EnvironmentSatisfaction</th>\n",
       "      <th>Gender</th>\n",
       "      <th>HourlyRate</th>\n",
       "      <th>JobInvolvement</th>\n",
       "      <th>JobLevel</th>\n",
       "      <th>JobRole</th>\n",
       "      <th>JobSatisfaction</th>\n",
       "      <th>MaritalStatus</th>\n",
       "      <th>MonthlyIncome</th>\n",
       "      <th>MonthlyRate</th>\n",
       "      <th>NumCompaniesWorked</th>\n",
       "      <th>Over18</th>\n",
       "      <th>OverTime</th>\n",
       "      <th>PercentSalaryHike</th>\n",
       "      <th>PerformanceRating</th>\n",
       "      <th>RelationshipSatisfaction</th>\n",
       "      <th>StandardHours</th>\n",
       "      <th>StockOptionLevel</th>\n",
       "      <th>TotalWorkingYears</th>\n",
       "      <th>TrainingTimesLastYear</th>\n",
       "      <th>WorkLifeBalance</th>\n",
       "      <th>YearsAtCompany</th>\n",
       "      <th>YearsInCurrentRole</th>\n",
       "      <th>YearsSinceLastPromotion</th>\n",
       "      <th>YearsWithCurrManager</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1114</th>\n",
       "      <td>2791</td>\n",
       "      <td>31</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>755</td>\n",
       "      <td>Sales</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Life Sciences</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Male</td>\n",
       "      <td>81</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>Sales Representative</td>\n",
       "      <td>4</td>\n",
       "      <td>Divorced</td>\n",
       "      <td>4678</td>\n",
       "      <td>9150</td>\n",
       "      <td>1</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>12</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1115</th>\n",
       "      <td>2792</td>\n",
       "      <td>40</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>654</td>\n",
       "      <td>Research &amp; Development</td>\n",
       "      <td>26</td>\n",
       "      <td>5</td>\n",
       "      <td>Medical</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Male</td>\n",
       "      <td>96</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>Laboratory Technician</td>\n",
       "      <td>4</td>\n",
       "      <td>Married</td>\n",
       "      <td>6220</td>\n",
       "      <td>6409</td>\n",
       "      <td>1</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1116</th>\n",
       "      <td>2793</td>\n",
       "      <td>42</td>\n",
       "      <td>Travel_Frequently</td>\n",
       "      <td>884</td>\n",
       "      <td>Research &amp; Development</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>Medical</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>Female</td>\n",
       "      <td>65</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>Healthcare Representative</td>\n",
       "      <td>1</td>\n",
       "      <td>Married</td>\n",
       "      <td>5238</td>\n",
       "      <td>6069</td>\n",
       "      <td>3</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>17</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1117</th>\n",
       "      <td>2794</td>\n",
       "      <td>25</td>\n",
       "      <td>Travel_Frequently</td>\n",
       "      <td>1469</td>\n",
       "      <td>Sales</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>Technical Degree</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Male</td>\n",
       "      <td>68</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>Sales Executive</td>\n",
       "      <td>4</td>\n",
       "      <td>Married</td>\n",
       "      <td>4799</td>\n",
       "      <td>17519</td>\n",
       "      <td>0</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>11</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1118</th>\n",
       "      <td>2795</td>\n",
       "      <td>42</td>\n",
       "      <td>Travel_Rarely</td>\n",
       "      <td>1234</td>\n",
       "      <td>Research &amp; Development</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>Life Sciences</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Female</td>\n",
       "      <td>95</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>Manager</td>\n",
       "      <td>4</td>\n",
       "      <td>Married</td>\n",
       "      <td>1706</td>\n",
       "      <td>13953</td>\n",
       "      <td>2</td>\n",
       "      <td>Y</td>\n",
       "      <td>No</td>\n",
       "      <td>11</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>80</td>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>12</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        id  Age     BusinessTravel  DailyRate              Department  \\\n",
       "1114  2791   31      Travel_Rarely        755                   Sales   \n",
       "1115  2792   40      Travel_Rarely        654  Research & Development   \n",
       "1116  2793   42  Travel_Frequently        884  Research & Development   \n",
       "1117  2794   25  Travel_Frequently       1469                   Sales   \n",
       "1118  2795   42      Travel_Rarely       1234  Research & Development   \n",
       "\n",
       "      DistanceFromHome  Education    EducationField  EmployeeCount  \\\n",
       "1114                 1          1     Life Sciences              1   \n",
       "1115                26          5           Medical              1   \n",
       "1116                 1          4           Medical              1   \n",
       "1117                 1          2  Technical Degree              1   \n",
       "1118                 2          4     Life Sciences              1   \n",
       "\n",
       "      EnvironmentSatisfaction  Gender  HourlyRate  JobInvolvement  JobLevel  \\\n",
       "1114                        3    Male          81               2         1   \n",
       "1115                        3    Male          96               2         2   \n",
       "1116                        2  Female          65               3         2   \n",
       "1117                        3    Male          68               2         2   \n",
       "1118                        3  Female          95               3         4   \n",
       "\n",
       "                        JobRole  JobSatisfaction MaritalStatus  MonthlyIncome  \\\n",
       "1114       Sales Representative                4      Divorced           4678   \n",
       "1115      Laboratory Technician                4       Married           6220   \n",
       "1116  Healthcare Representative                1       Married           5238   \n",
       "1117            Sales Executive                4       Married           4799   \n",
       "1118                    Manager                4       Married           1706   \n",
       "\n",
       "      MonthlyRate  NumCompaniesWorked Over18 OverTime  PercentSalaryHike  \\\n",
       "1114         9150                   1      Y       No                 12   \n",
       "1115         6409                   1      Y       No                 14   \n",
       "1116         6069                   3      Y       No                 17   \n",
       "1117        17519                   0      Y       No                 11   \n",
       "1118        13953                   2      Y       No                 11   \n",
       "\n",
       "      PerformanceRating  RelationshipSatisfaction  StandardHours  \\\n",
       "1114                  3                         3             80   \n",
       "1115                  3                         4             80   \n",
       "1116                  3                         1             80   \n",
       "1117                  3                         4             80   \n",
       "1118                  3                         3             80   \n",
       "\n",
       "      StockOptionLevel  TotalWorkingYears  TrainingTimesLastYear  \\\n",
       "1114                 1                  1                      1   \n",
       "1115                 1                 20                      4   \n",
       "1116                 1                 10                      2   \n",
       "1117                 1                  5                      2   \n",
       "1118                 1                 23                      2   \n",
       "\n",
       "      WorkLifeBalance  YearsAtCompany  YearsInCurrentRole  \\\n",
       "1114                3               1                   0   \n",
       "1115                3              20                  10   \n",
       "1116                2               5                   3   \n",
       "1117                3               4                   2   \n",
       "1118                3              12                   7   \n",
       "\n",
       "      YearsSinceLastPromotion  YearsWithCurrManager  \n",
       "1114                        0                     0  \n",
       "1115                        1                     8  \n",
       "1116                        0                     2  \n",
       "1117                        1                     3  \n",
       "1118                        4                     9  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "8f37374f-b0f5-48e0-ba63-ffff70984143",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1677 entries, 0 to 1676\n",
      "Data columns (total 35 columns):\n",
      " #   Column                    Non-Null Count  Dtype \n",
      "---  ------                    --------------  ----- \n",
      " 0   id                        1677 non-null   int64 \n",
      " 1   Age                       1677 non-null   int64 \n",
      " 2   BusinessTravel            1677 non-null   object\n",
      " 3   DailyRate                 1677 non-null   int64 \n",
      " 4   Department                1677 non-null   object\n",
      " 5   DistanceFromHome          1677 non-null   int64 \n",
      " 6   Education                 1677 non-null   int64 \n",
      " 7   EducationField            1677 non-null   object\n",
      " 8   EmployeeCount             1677 non-null   int64 \n",
      " 9   EnvironmentSatisfaction   1677 non-null   int64 \n",
      " 10  Gender                    1677 non-null   object\n",
      " 11  HourlyRate                1677 non-null   int64 \n",
      " 12  JobInvolvement            1677 non-null   int64 \n",
      " 13  JobLevel                  1677 non-null   int64 \n",
      " 14  JobRole                   1677 non-null   object\n",
      " 15  JobSatisfaction           1677 non-null   int64 \n",
      " 16  MaritalStatus             1677 non-null   object\n",
      " 17  MonthlyIncome             1677 non-null   int64 \n",
      " 18  MonthlyRate               1677 non-null   int64 \n",
      " 19  NumCompaniesWorked        1677 non-null   int64 \n",
      " 20  Over18                    1677 non-null   object\n",
      " 21  OverTime                  1677 non-null   object\n",
      " 22  PercentSalaryHike         1677 non-null   int64 \n",
      " 23  PerformanceRating         1677 non-null   int64 \n",
      " 24  RelationshipSatisfaction  1677 non-null   int64 \n",
      " 25  StandardHours             1677 non-null   int64 \n",
      " 26  StockOptionLevel          1677 non-null   int64 \n",
      " 27  TotalWorkingYears         1677 non-null   int64 \n",
      " 28  TrainingTimesLastYear     1677 non-null   int64 \n",
      " 29  WorkLifeBalance           1677 non-null   int64 \n",
      " 30  YearsAtCompany            1677 non-null   int64 \n",
      " 31  YearsInCurrentRole        1677 non-null   int64 \n",
      " 32  YearsSinceLastPromotion   1677 non-null   int64 \n",
      " 33  YearsWithCurrManager      1677 non-null   int64 \n",
      " 34  Attrition                 1677 non-null   int64 \n",
      "dtypes: int64(27), object(8)\n",
      "memory usage: 458.7+ KB\n"
     ]
    }
   ],
   "source": [
    "df1.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "2a92a2c0-4c8b-409a-82ec-06a545da2aea",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1119 entries, 0 to 1118\n",
      "Data columns (total 34 columns):\n",
      " #   Column                    Non-Null Count  Dtype \n",
      "---  ------                    --------------  ----- \n",
      " 0   id                        1119 non-null   int64 \n",
      " 1   Age                       1119 non-null   int64 \n",
      " 2   BusinessTravel            1119 non-null   object\n",
      " 3   DailyRate                 1119 non-null   int64 \n",
      " 4   Department                1119 non-null   object\n",
      " 5   DistanceFromHome          1119 non-null   int64 \n",
      " 6   Education                 1119 non-null   int64 \n",
      " 7   EducationField            1119 non-null   object\n",
      " 8   EmployeeCount             1119 non-null   int64 \n",
      " 9   EnvironmentSatisfaction   1119 non-null   int64 \n",
      " 10  Gender                    1119 non-null   object\n",
      " 11  HourlyRate                1119 non-null   int64 \n",
      " 12  JobInvolvement            1119 non-null   int64 \n",
      " 13  JobLevel                  1119 non-null   int64 \n",
      " 14  JobRole                   1119 non-null   object\n",
      " 15  JobSatisfaction           1119 non-null   int64 \n",
      " 16  MaritalStatus             1119 non-null   object\n",
      " 17  MonthlyIncome             1119 non-null   int64 \n",
      " 18  MonthlyRate               1119 non-null   int64 \n",
      " 19  NumCompaniesWorked        1119 non-null   int64 \n",
      " 20  Over18                    1119 non-null   object\n",
      " 21  OverTime                  1119 non-null   object\n",
      " 22  PercentSalaryHike         1119 non-null   int64 \n",
      " 23  PerformanceRating         1119 non-null   int64 \n",
      " 24  RelationshipSatisfaction  1119 non-null   int64 \n",
      " 25  StandardHours             1119 non-null   int64 \n",
      " 26  StockOptionLevel          1119 non-null   int64 \n",
      " 27  TotalWorkingYears         1119 non-null   int64 \n",
      " 28  TrainingTimesLastYear     1119 non-null   int64 \n",
      " 29  WorkLifeBalance           1119 non-null   int64 \n",
      " 30  YearsAtCompany            1119 non-null   int64 \n",
      " 31  YearsInCurrentRole        1119 non-null   int64 \n",
      " 32  YearsSinceLastPromotion   1119 non-null   int64 \n",
      " 33  YearsWithCurrManager      1119 non-null   int64 \n",
      "dtypes: int64(26), object(8)\n",
      "memory usage: 297.4+ KB\n"
     ]
    }
   ],
   "source": [
    "df2.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8bd7538c-c2a9-46e8-b799-4fe3fa3c7e25",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id                          0\n",
       "Age                         0\n",
       "BusinessTravel              0\n",
       "DailyRate                   0\n",
       "Department                  0\n",
       "DistanceFromHome            0\n",
       "Education                   0\n",
       "EducationField              0\n",
       "EmployeeCount               0\n",
       "EnvironmentSatisfaction     0\n",
       "Gender                      0\n",
       "HourlyRate                  0\n",
       "JobInvolvement              0\n",
       "JobLevel                    0\n",
       "JobRole                     0\n",
       "JobSatisfaction             0\n",
       "MaritalStatus               0\n",
       "MonthlyIncome               0\n",
       "MonthlyRate                 0\n",
       "NumCompaniesWorked          0\n",
       "Over18                      0\n",
       "OverTime                    0\n",
       "PercentSalaryHike           0\n",
       "PerformanceRating           0\n",
       "RelationshipSatisfaction    0\n",
       "StandardHours               0\n",
       "StockOptionLevel            0\n",
       "TotalWorkingYears           0\n",
       "TrainingTimesLastYear       0\n",
       "WorkLifeBalance             0\n",
       "YearsAtCompany              0\n",
       "YearsInCurrentRole          0\n",
       "YearsSinceLastPromotion     0\n",
       "YearsWithCurrManager        0\n",
       "Attrition                   0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "98a4edf1-8bf6-494a-a7ef-c3e4c5ca55a2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id                          0\n",
       "Age                         0\n",
       "BusinessTravel              0\n",
       "DailyRate                   0\n",
       "Department                  0\n",
       "DistanceFromHome            0\n",
       "Education                   0\n",
       "EducationField              0\n",
       "EmployeeCount               0\n",
       "EnvironmentSatisfaction     0\n",
       "Gender                      0\n",
       "HourlyRate                  0\n",
       "JobInvolvement              0\n",
       "JobLevel                    0\n",
       "JobRole                     0\n",
       "JobSatisfaction             0\n",
       "MaritalStatus               0\n",
       "MonthlyIncome               0\n",
       "MonthlyRate                 0\n",
       "NumCompaniesWorked          0\n",
       "Over18                      0\n",
       "OverTime                    0\n",
       "PercentSalaryHike           0\n",
       "PerformanceRating           0\n",
       "RelationshipSatisfaction    0\n",
       "StandardHours               0\n",
       "StockOptionLevel            0\n",
       "TotalWorkingYears           0\n",
       "TrainingTimesLastYear       0\n",
       "WorkLifeBalance             0\n",
       "YearsAtCompany              0\n",
       "YearsInCurrentRole          0\n",
       "YearsSinceLastPromotion     0\n",
       "YearsWithCurrManager        0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "a61af16a-ff1c-41b4-9d80-37e976e97697",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1677, 35)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "29a27315-0dde-4e3b-bf47-5e1771180085",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1119, 34)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "b9dee849-5184-422d-ba67-e76722918722",
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.concat([df1,df2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "0c75df7b-9eb9-4eaf-908a-4dfb59c9186e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12,8))\n",
    "sns.heatmap(df1.corr(numeric_only=True)[['Attrition']].sort_values('Attrition', ascending=False),annot=True, cmap='coolwarm')\n",
    "plt.title('Correlation with Attrition')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "026eb969-b693-45d3-bb4a-98a198520576",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12,10))\n",
    "sns.heatmap(\n",
    "    df1.drop(columns=['Attrition']).corr(numeric_only=True),\n",
    "    cmap='coolwarm'\n",
    ")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d440cda-11e8-47f2-b5fe-7162bdea3c8c",
   "metadata": {},
   "source": [
    "The correlation analysis shows strong relationships among seniority-related features such as JobLevel, MonthlyIncome, and TotalWorkingYears, indicating feature redundancy. Most other features have weak linear correlations, suggesting that non-linear models are more suitable for this task."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "caee4083-0786-4a91-84d2-30edae29563b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Attrition\n",
       "0    1477\n",
       "1     200\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1['Attrition'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "3c0f6c24-7af5-4557-9057-6942d9f30e2d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(5,4))\n",
    "sns.countplot(x='Attrition', data=df1)\n",
    "plt.title('Target Distribution (Attrition)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4cd2d1cf-9fd0-4719-a698-b95955d77f21",
   "metadata": {},
   "source": [
    "The target variable is imbalanced, with approximately 12% attrition and 88% non-attrition."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "266ecdf0-b55b-4973-bd0a-ee084c2128ab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Attrition</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JobRole</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Healthcare Representative</th>\n",
       "      <td>0.935484</td>\n",
       "      <td>0.064516</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Human Resources</th>\n",
       "      <td>0.794118</td>\n",
       "      <td>0.205882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Laboratory Technician</th>\n",
       "      <td>0.817365</td>\n",
       "      <td>0.182635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Manager</th>\n",
       "      <td>0.918919</td>\n",
       "      <td>0.081081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Manufacturing Director</th>\n",
       "      <td>0.974490</td>\n",
       "      <td>0.025510</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Research Director</th>\n",
       "      <td>0.971831</td>\n",
       "      <td>0.028169</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Research Scientist</th>\n",
       "      <td>0.877907</td>\n",
       "      <td>0.122093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sales Executive</th>\n",
       "      <td>0.892958</td>\n",
       "      <td>0.107042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sales Representative</th>\n",
       "      <td>0.662338</td>\n",
       "      <td>0.337662</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Attrition                         0         1\n",
       "JobRole                                      \n",
       "Healthcare Representative  0.935484  0.064516\n",
       "Human Resources            0.794118  0.205882\n",
       "Laboratory Technician      0.817365  0.182635\n",
       "Manager                    0.918919  0.081081\n",
       "Manufacturing Director     0.974490  0.025510\n",
       "Research Director          0.971831  0.028169\n",
       "Research Scientist         0.877907  0.122093\n",
       "Sales Executive            0.892958  0.107042\n",
       "Sales Representative       0.662338  0.337662"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(df1['OverTime'], df1['Attrition'], normalize='index')\n",
    "pd.crosstab(df1['JobRole'], df1['Attrition'], normalize='index')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95cc3d58-5d33-495c-96ca-06a8f083ca73",
   "metadata": {},
   "source": [
    "Attrition rates vary significantly across job roles, with Sales Representative and Human Resources showing notably higher attrition rates compared to managerial and director roles."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "17d4ef42-5f94-4be6-934f-738f0e14d026",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Attrition\n",
       "0    36.540961\n",
       "1    32.315000\n",
       "Name: Age, dtype: float64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.groupby('Attrition')['MonthlyIncome'].mean()\n",
    "df1.groupby('Attrition')['Age'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "c0689ac1-4027-4814-b905-e0afa856c510",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "EmployeeCount                  1\n",
       "Over18                         1\n",
       "StandardHours                  1\n",
       "Gender                         2\n",
       "OverTime                       2\n",
       "PerformanceRating              2\n",
       "Attrition                      2\n",
       "BusinessTravel                 3\n",
       "MaritalStatus                  3\n",
       "Department                     3\n",
       "RelationshipSatisfaction       4\n",
       "JobSatisfaction                4\n",
       "WorkLifeBalance                4\n",
       "StockOptionLevel               5\n",
       "JobInvolvement                 5\n",
       "EnvironmentSatisfaction        5\n",
       "JobLevel                       6\n",
       "EducationField                 6\n",
       "Education                      6\n",
       "TrainingTimesLastYear          7\n",
       "JobRole                        9\n",
       "NumCompaniesWorked            11\n",
       "PercentSalaryHike             15\n",
       "YearsSinceLastPromotion       16\n",
       "YearsWithCurrManager          18\n",
       "YearsInCurrentRole            19\n",
       "DistanceFromHome              29\n",
       "YearsAtCompany                37\n",
       "TotalWorkingYears             41\n",
       "Age                           43\n",
       "HourlyRate                    71\n",
       "DailyRate                    734\n",
       "MonthlyRate                 1127\n",
       "MonthlyIncome               1140\n",
       "id                          2796\n",
       "dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.nunique().sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "1981c919-922d-4497-9d80-0a119d6ec09c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "EmployeeCount\n",
       "1    2796\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['EmployeeCount'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "e25e9684-be16-44bc-89d4-0bb679a3a0d9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Over18\n",
       "Y    2796\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Over18'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "33bc19a0-bdf4-4c3f-8e3e-e3bc7521e1b2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "StandardHours\n",
       "80    2796\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['StandardHours'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "56547481-3d35-4a07-bfca-697992b5ef27",
   "metadata": {},
   "outputs": [],
   "source": [
    "df=df.drop(columns=['EmployeeCount','Over18','StandardHours','id'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "1b5c3e87-1ebf-4027-843d-951d3dd464fa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Index: 2796 entries, 0 to 1118\n",
      "Data columns (total 31 columns):\n",
      " #   Column                    Non-Null Count  Dtype  \n",
      "---  ------                    --------------  -----  \n",
      " 0   Age                       2796 non-null   int64  \n",
      " 1   BusinessTravel            2796 non-null   object \n",
      " 2   DailyRate                 2796 non-null   int64  \n",
      " 3   Department                2796 non-null   object \n",
      " 4   DistanceFromHome          2796 non-null   int64  \n",
      " 5   Education                 2796 non-null   int64  \n",
      " 6   EducationField            2796 non-null   object \n",
      " 7   EnvironmentSatisfaction   2796 non-null   int64  \n",
      " 8   Gender                    2796 non-null   object \n",
      " 9   HourlyRate                2796 non-null   int64  \n",
      " 10  JobInvolvement            2796 non-null   int64  \n",
      " 11  JobLevel                  2796 non-null   int64  \n",
      " 12  JobRole                   2796 non-null   object \n",
      " 13  JobSatisfaction           2796 non-null   int64  \n",
      " 14  MaritalStatus             2796 non-null   object \n",
      " 15  MonthlyIncome             2796 non-null   int64  \n",
      " 16  MonthlyRate               2796 non-null   int64  \n",
      " 17  NumCompaniesWorked        2796 non-null   int64  \n",
      " 18  OverTime                  2796 non-null   object \n",
      " 19  PercentSalaryHike         2796 non-null   int64  \n",
      " 20  PerformanceRating         2796 non-null   int64  \n",
      " 21  RelationshipSatisfaction  2796 non-null   int64  \n",
      " 22  StockOptionLevel          2796 non-null   int64  \n",
      " 23  TotalWorkingYears         2796 non-null   int64  \n",
      " 24  TrainingTimesLastYear     2796 non-null   int64  \n",
      " 25  WorkLifeBalance           2796 non-null   int64  \n",
      " 26  YearsAtCompany            2796 non-null   int64  \n",
      " 27  YearsInCurrentRole        2796 non-null   int64  \n",
      " 28  YearsSinceLastPromotion   2796 non-null   int64  \n",
      " 29  YearsWithCurrManager      2796 non-null   int64  \n",
      " 30  Attrition                 1677 non-null   float64\n",
      "dtypes: float64(1), int64(23), object(7)\n",
      "memory usage: 699.0+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1e221a68-523e-4fdd-a05e-1dfc988f78df",
   "metadata": {},
   "source": [
    "### Feature Engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "e33e0c01-cb59-4ac3-afc9-de54a759e5ab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OverTime\n",
       "No     2129\n",
       "Yes     667\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['OverTime'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "a91c8f2e-453b-4ad3-b862-c4d891ed39af",
   "metadata": {},
   "outputs": [],
   "source": [
    "d={'No':0, 'Yes':1}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "6b1f6a0b-e1af-4687-b158-e2124a34d79d",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['OverTime']=df['OverTime'].map(d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "d516d616-4d8a-4941-88eb-775ab934e13b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MaritalStatus\n",
       "Married     1297\n",
       "Single       971\n",
       "Divorced     528\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['MaritalStatus'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "6bba1d2c-2792-430e-b9bd-6ca22685edd7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "JobRole\n",
       "Sales Executive              595\n",
       "Research Scientist           574\n",
       "Laboratory Technician        564\n",
       "Manufacturing Director       313\n",
       "Healthcare Representative    267\n",
       "Manager                      182\n",
       "Sales Representative         121\n",
       "Research Director            117\n",
       "Human Resources               63\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['JobRole'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "55a4b190-3982-4233-ae09-7fdaa3857ae9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender\n",
       "Male      1754\n",
       "Female    1042\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Gender'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "60f3056b-7e1e-427d-95a2-c0a029046b8d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "EducationField\n",
       "Life Sciences       1294\n",
       "Medical              894\n",
       "Marketing            254\n",
       "Technical Degree     202\n",
       "Other                127\n",
       "Human Resources       25\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['EducationField'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "2aea5a2c-6c87-4c7b-94e0-d536c12f416f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Department\n",
       "Research & Development    1944\n",
       "Sales                      777\n",
       "Human Resources             75\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Department'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "af14e463-df7f-48d5-a842-58a572c9f551",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "BusinessTravel\n",
       "Travel_Rarely        2089\n",
       "Travel_Frequently     473\n",
       "Non-Travel            234\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['BusinessTravel'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "80ea9e59-a388-4e5e-a420-1dc1ac9a5754",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Company experience share\n",
    "df['time_experience']=df['YearsAtCompany'] / (df['TotalWorkingYears'] + 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "0194db74-8abd-4375-903e-bcf18bd7ed05",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Time since last promotion\n",
    "df['no_promotion']=df['YearsAtCompany'] - df['YearsSinceLastPromotion']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "d1ae3085-64ed-4731-8d6d-9433187e09ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compensation relative to level\n",
    "df['income_experience']=df['MonthlyIncome'] / (df['JobLevel'] + 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "5ca799a6-99aa-4c84-a870-2b3b1a80b62d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Manager relationship stability\n",
    "df['manager_stability']=df['YearsWithCurrManager'] / (df['YearsAtCompany'] + 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "1ed165fc-d3a4-4b56-9939-ae2175ca26d5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Index: 2796 entries, 0 to 1118\n",
      "Data columns (total 35 columns):\n",
      " #   Column                    Non-Null Count  Dtype  \n",
      "---  ------                    --------------  -----  \n",
      " 0   Age                       2796 non-null   int64  \n",
      " 1   BusinessTravel            2796 non-null   object \n",
      " 2   DailyRate                 2796 non-null   int64  \n",
      " 3   Department                2796 non-null   object \n",
      " 4   DistanceFromHome          2796 non-null   int64  \n",
      " 5   Education                 2796 non-null   int64  \n",
      " 6   EducationField            2796 non-null   object \n",
      " 7   EnvironmentSatisfaction   2796 non-null   int64  \n",
      " 8   Gender                    2796 non-null   object \n",
      " 9   HourlyRate                2796 non-null   int64  \n",
      " 10  JobInvolvement            2796 non-null   int64  \n",
      " 11  JobLevel                  2796 non-null   int64  \n",
      " 12  JobRole                   2796 non-null   object \n",
      " 13  JobSatisfaction           2796 non-null   int64  \n",
      " 14  MaritalStatus             2796 non-null   object \n",
      " 15  MonthlyIncome             2796 non-null   int64  \n",
      " 16  MonthlyRate               2796 non-null   int64  \n",
      " 17  NumCompaniesWorked        2796 non-null   int64  \n",
      " 18  OverTime                  2796 non-null   int64  \n",
      " 19  PercentSalaryHike         2796 non-null   int64  \n",
      " 20  PerformanceRating         2796 non-null   int64  \n",
      " 21  RelationshipSatisfaction  2796 non-null   int64  \n",
      " 22  StockOptionLevel          2796 non-null   int64  \n",
      " 23  TotalWorkingYears         2796 non-null   int64  \n",
      " 24  TrainingTimesLastYear     2796 non-null   int64  \n",
      " 25  WorkLifeBalance           2796 non-null   int64  \n",
      " 26  YearsAtCompany            2796 non-null   int64  \n",
      " 27  YearsInCurrentRole        2796 non-null   int64  \n",
      " 28  YearsSinceLastPromotion   2796 non-null   int64  \n",
      " 29  YearsWithCurrManager      2796 non-null   int64  \n",
      " 30  Attrition                 1677 non-null   float64\n",
      " 31  time_experience           2796 non-null   float64\n",
      " 32  no_promotion              2796 non-null   int64  \n",
      " 33  income_experience         2796 non-null   float64\n",
      " 34  manager_stability         2796 non-null   float64\n",
      "dtypes: float64(4), int64(25), object(6)\n",
      "memory usage: 786.4+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "912ea73f-4958-428e-9dfe-71fdb7444dc4",
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.get_dummies(df,drop_first=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b1ad21b-856c-4860-97b1-3cfbae846c0a",
   "metadata": {},
   "source": [
    "### Modelling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "c9239f0a-aa13-4a59-b330-c9455a0b03fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "train=df[:1677]\n",
    "test=df[1677:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "617f786e-7395-4cc8-9da7-267fe714f4ad",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>DailyRate</th>\n",
       "      <th>DistanceFromHome</th>\n",
       "      <th>Education</th>\n",
       "      <th>EnvironmentSatisfaction</th>\n",
       "      <th>HourlyRate</th>\n",
       "      <th>JobInvolvement</th>\n",
       "      <th>JobLevel</th>\n",
       "      <th>JobSatisfaction</th>\n",
       "      <th>MonthlyIncome</th>\n",
       "      <th>MonthlyRate</th>\n",
       "      <th>NumCompaniesWorked</th>\n",
       "      <th>OverTime</th>\n",
       "      <th>PercentSalaryHike</th>\n",
       "      <th>PerformanceRating</th>\n",
       "      <th>RelationshipSatisfaction</th>\n",
       "      <th>StockOptionLevel</th>\n",
       "      <th>TotalWorkingYears</th>\n",
       "      <th>TrainingTimesLastYear</th>\n",
       "      <th>WorkLifeBalance</th>\n",
       "      <th>YearsAtCompany</th>\n",
       "      <th>YearsInCurrentRole</th>\n",
       "      <th>YearsSinceLastPromotion</th>\n",
       "      <th>YearsWithCurrManager</th>\n",
       "      <th>Attrition</th>\n",
       "      <th>time_experience</th>\n",
       "      <th>no_promotion</th>\n",
       "      <th>income_experience</th>\n",
       "      <th>manager_stability</th>\n",
       "      <th>BusinessTravel_Travel_Frequently</th>\n",
       "      <th>BusinessTravel_Travel_Rarely</th>\n",
       "      <th>Department_Research &amp; Development</th>\n",
       "      <th>Department_Sales</th>\n",
       "      <th>EducationField_Life Sciences</th>\n",
       "      <th>EducationField_Marketing</th>\n",
       "      <th>EducationField_Medical</th>\n",
       "      <th>EducationField_Other</th>\n",
       "      <th>EducationField_Technical Degree</th>\n",
       "      <th>Gender_Male</th>\n",
       "      <th>JobRole_Human Resources</th>\n",
       "      <th>JobRole_Laboratory Technician</th>\n",
       "      <th>JobRole_Manager</th>\n",
       "      <th>JobRole_Manufacturing Director</th>\n",
       "      <th>JobRole_Research Director</th>\n",
       "      <th>JobRole_Research Scientist</th>\n",
       "      <th>JobRole_Sales Executive</th>\n",
       "      <th>JobRole_Sales Representative</th>\n",
       "      <th>MaritalStatus_Married</th>\n",
       "      <th>MaritalStatus_Single</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1672</th>\n",
       "      <td>30</td>\n",
       "      <td>945</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>73</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>8722</td>\n",
       "      <td>14255</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>19</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.909091</td>\n",
       "      <td>10</td>\n",
       "      <td>2180.5</td>\n",
       "      <td>0.727273</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1673</th>\n",
       "      <td>32</td>\n",
       "      <td>1303</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>48</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3544</td>\n",
       "      <td>15972</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>19</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.363636</td>\n",
       "      <td>3</td>\n",
       "      <td>1772.0</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1674</th>\n",
       "      <td>29</td>\n",
       "      <td>1184</td>\n",
       "      <td>24</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>36</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2804</td>\n",
       "      <td>15322</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>1</td>\n",
       "      <td>1402.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1675</th>\n",
       "      <td>36</td>\n",
       "      <td>441</td>\n",
       "      <td>9</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>48</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>5406</td>\n",
       "      <td>4051</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>21</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.909091</td>\n",
       "      <td>10</td>\n",
       "      <td>1802.0</td>\n",
       "      <td>0.727273</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1676</th>\n",
       "      <td>36</td>\n",
       "      <td>1141</td>\n",
       "      <td>20</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>35</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2593</td>\n",
       "      <td>17381</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>19</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.727273</td>\n",
       "      <td>1</td>\n",
       "      <td>1296.5</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Age  DailyRate  DistanceFromHome  Education  EnvironmentSatisfaction  \\\n",
       "1672   30        945                 1          3                        4   \n",
       "1673   32       1303                 2          3                        1   \n",
       "1674   29       1184                24          3                        2   \n",
       "1675   36        441                 9          2                        2   \n",
       "1676   36       1141                20          3                        3   \n",
       "\n",
       "      HourlyRate  JobInvolvement  JobLevel  JobSatisfaction  MonthlyIncome  \\\n",
       "1672          73               3         3                3           8722   \n",
       "1673          48               3         1                2           3544   \n",
       "1674          36               2         1                1           2804   \n",
       "1675          48               4         2                3           5406   \n",
       "1676          35               3         1                3           2593   \n",
       "\n",
       "      MonthlyRate  NumCompaniesWorked  OverTime  PercentSalaryHike  \\\n",
       "1672        14255                   1         0                 19   \n",
       "1673        15972                   4         0                 19   \n",
       "1674        15322                   1         1                 11   \n",
       "1675         4051                   1         0                 21   \n",
       "1676        17381                   0         0                 19   \n",
       "\n",
       "      PerformanceRating  RelationshipSatisfaction  StockOptionLevel  \\\n",
       "1672                  3                         2                 0   \n",
       "1673                  3                         4                 1   \n",
       "1674                  3                         3                 0   \n",
       "1675                  4                         3                 2   \n",
       "1676                  3                         4                 0   \n",
       "\n",
       "      TotalWorkingYears  TrainingTimesLastYear  WorkLifeBalance  \\\n",
       "1672                 10                      2                4   \n",
       "1673                 10                      3                4   \n",
       "1674                  1                      2                3   \n",
       "1675                 10                      3                2   \n",
       "1676                 10                      3                2   \n",
       "\n",
       "      YearsAtCompany  YearsInCurrentRole  YearsSinceLastPromotion  \\\n",
       "1672              10                   0                        0   \n",
       "1673               4                   2                        1   \n",
       "1674               1                   0                        0   \n",
       "1675              10                   3                        0   \n",
       "1676               8                   2                        7   \n",
       "\n",
       "      YearsWithCurrManager  Attrition  time_experience  no_promotion  \\\n",
       "1672                     8        0.0         0.909091            10   \n",
       "1673                     3        0.0         0.363636             3   \n",
       "1674                     0        1.0         0.500000             1   \n",
       "1675                     8        0.0         0.909091            10   \n",
       "1676                     3        0.0         0.727273             1   \n",
       "\n",
       "      income_experience  manager_stability  BusinessTravel_Travel_Frequently  \\\n",
       "1672             2180.5           0.727273                             False   \n",
       "1673             1772.0           0.600000                             False   \n",
       "1674             1402.0           0.000000                              True   \n",
       "1675             1802.0           0.727273                             False   \n",
       "1676             1296.5           0.333333                             False   \n",
       "\n",
       "      BusinessTravel_Travel_Rarely  Department_Research & Development  \\\n",
       "1672                          True                              False   \n",
       "1673                          True                               True   \n",
       "1674                         False                              False   \n",
       "1675                          True                              False   \n",
       "1676                          True                               True   \n",
       "\n",
       "      Department_Sales  EducationField_Life Sciences  \\\n",
       "1672              True                          True   \n",
       "1673             False                          True   \n",
       "1674             False                         False   \n",
       "1675              True                         False   \n",
       "1676             False                          True   \n",
       "\n",
       "      EducationField_Marketing  EducationField_Medical  EducationField_Other  \\\n",
       "1672                     False                   False                 False   \n",
       "1673                     False                   False                 False   \n",
       "1674                     False                   False                 False   \n",
       "1675                      True                   False                 False   \n",
       "1676                     False                   False                 False   \n",
       "\n",
       "      EducationField_Technical Degree  Gender_Male  JobRole_Human Resources  \\\n",
       "1672                            False        False                    False   \n",
       "1673                            False         True                    False   \n",
       "1674                            False         True                     True   \n",
       "1675                            False         True                    False   \n",
       "1676                            False        False                    False   \n",
       "\n",
       "      JobRole_Laboratory Technician  JobRole_Manager  \\\n",
       "1672                          False            False   \n",
       "1673                          False            False   \n",
       "1674                          False            False   \n",
       "1675                          False            False   \n",
       "1676                           True            False   \n",
       "\n",
       "      JobRole_Manufacturing Director  JobRole_Research Director  \\\n",
       "1672                           False                      False   \n",
       "1673                           False                      False   \n",
       "1674                           False                      False   \n",
       "1675                           False                      False   \n",
       "1676                           False                      False   \n",
       "\n",
       "      JobRole_Research Scientist  JobRole_Sales Executive  \\\n",
       "1672                       False                     True   \n",
       "1673                        True                    False   \n",
       "1674                       False                    False   \n",
       "1675                       False                     True   \n",
       "1676                       False                    False   \n",
       "\n",
       "      JobRole_Sales Representative  MaritalStatus_Married  \\\n",
       "1672                         False                  False   \n",
       "1673                         False                   True   \n",
       "1674                         False                   True   \n",
       "1675                         False                  False   \n",
       "1676                         False                  False   \n",
       "\n",
       "      MaritalStatus_Single  \n",
       "1672                  True  \n",
       "1673                 False  \n",
       "1674                 False  \n",
       "1675                 False  \n",
       "1676                  True  "
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "ba95b8a5-1188-4724-a55a-ad361c33413f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>DailyRate</th>\n",
       "      <th>DistanceFromHome</th>\n",
       "      <th>Education</th>\n",
       "      <th>EnvironmentSatisfaction</th>\n",
       "      <th>HourlyRate</th>\n",
       "      <th>JobInvolvement</th>\n",
       "      <th>JobLevel</th>\n",
       "      <th>JobSatisfaction</th>\n",
       "      <th>MonthlyIncome</th>\n",
       "      <th>MonthlyRate</th>\n",
       "      <th>NumCompaniesWorked</th>\n",
       "      <th>OverTime</th>\n",
       "      <th>PercentSalaryHike</th>\n",
       "      <th>PerformanceRating</th>\n",
       "      <th>RelationshipSatisfaction</th>\n",
       "      <th>StockOptionLevel</th>\n",
       "      <th>TotalWorkingYears</th>\n",
       "      <th>TrainingTimesLastYear</th>\n",
       "      <th>WorkLifeBalance</th>\n",
       "      <th>YearsAtCompany</th>\n",
       "      <th>YearsInCurrentRole</th>\n",
       "      <th>YearsSinceLastPromotion</th>\n",
       "      <th>YearsWithCurrManager</th>\n",
       "      <th>Attrition</th>\n",
       "      <th>time_experience</th>\n",
       "      <th>no_promotion</th>\n",
       "      <th>income_experience</th>\n",
       "      <th>manager_stability</th>\n",
       "      <th>BusinessTravel_Travel_Frequently</th>\n",
       "      <th>BusinessTravel_Travel_Rarely</th>\n",
       "      <th>Department_Research &amp; Development</th>\n",
       "      <th>Department_Sales</th>\n",
       "      <th>EducationField_Life Sciences</th>\n",
       "      <th>EducationField_Marketing</th>\n",
       "      <th>EducationField_Medical</th>\n",
       "      <th>EducationField_Other</th>\n",
       "      <th>EducationField_Technical Degree</th>\n",
       "      <th>Gender_Male</th>\n",
       "      <th>JobRole_Human Resources</th>\n",
       "      <th>JobRole_Laboratory Technician</th>\n",
       "      <th>JobRole_Manager</th>\n",
       "      <th>JobRole_Manufacturing Director</th>\n",
       "      <th>JobRole_Research Director</th>\n",
       "      <th>JobRole_Research Scientist</th>\n",
       "      <th>JobRole_Sales Executive</th>\n",
       "      <th>JobRole_Sales Representative</th>\n",
       "      <th>MaritalStatus_Married</th>\n",
       "      <th>MaritalStatus_Single</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>19</td>\n",
       "      <td>992</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>43</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2318</td>\n",
       "      <td>17778</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>1</td>\n",
       "      <td>1159.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>45</td>\n",
       "      <td>1136</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>67</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>5486</td>\n",
       "      <td>12421</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>12</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0</td>\n",
       "      <td>1828.666667</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>37</td>\n",
       "      <td>155</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>41</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2741</td>\n",
       "      <td>23577</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>13</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>6</td>\n",
       "      <td>1370.500000</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>32</td>\n",
       "      <td>688</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>89</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>5228</td>\n",
       "      <td>20364</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>14</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>14</td>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "      <td>8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.933333</td>\n",
       "      <td>3</td>\n",
       "      <td>1742.666667</td>\n",
       "      <td>0.533333</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>29</td>\n",
       "      <td>464</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>79</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>1223</td>\n",
       "      <td>15178</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>1</td>\n",
       "      <td>611.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Age  DailyRate  DistanceFromHome  Education  EnvironmentSatisfaction  \\\n",
       "0   19        992                 1          1                        4   \n",
       "1   45       1136                 4          4                        3   \n",
       "2   37        155                13          3                        4   \n",
       "3   32        688                 1          4                        3   \n",
       "4   29        464                 9          1                        3   \n",
       "\n",
       "   HourlyRate  JobInvolvement  JobLevel  JobSatisfaction  MonthlyIncome  \\\n",
       "0          43               3         1                3           2318   \n",
       "1          67               3         2                1           5486   \n",
       "2          41               3         1                4           2741   \n",
       "3          89               2         2                3           5228   \n",
       "4          79               3         1                4           1223   \n",
       "\n",
       "   MonthlyRate  NumCompaniesWorked  OverTime  PercentSalaryHike  \\\n",
       "0        17778                   1         0                 12   \n",
       "1        12421                   6         1                 12   \n",
       "2        23577                   4         1                 13   \n",
       "3        20364                   1         0                 13   \n",
       "4        15178                   1         0                 14   \n",
       "\n",
       "   PerformanceRating  RelationshipSatisfaction  StockOptionLevel  \\\n",
       "0                  3                         4                 0   \n",
       "1                  3                         3                 1   \n",
       "2                  3                         2                 2   \n",
       "3                  3                         3                 0   \n",
       "4                  3                         1                 0   \n",
       "\n",
       "   TotalWorkingYears  TrainingTimesLastYear  WorkLifeBalance  YearsAtCompany  \\\n",
       "0                  1                      2                2               1   \n",
       "1                  7                      3                3               2   \n",
       "2                 13                      2                2               7   \n",
       "3                 14                      2                2              14   \n",
       "4                  1                      5                3               1   \n",
       "\n",
       "   YearsInCurrentRole  YearsSinceLastPromotion  YearsWithCurrManager  \\\n",
       "0                   0                        0                     0   \n",
       "1                   2                        2                     2   \n",
       "2                   7                        1                     7   \n",
       "3                  10                       11                     8   \n",
       "4                   0                        0                     0   \n",
       "\n",
       "   Attrition  time_experience  no_promotion  income_experience  \\\n",
       "0        NaN         0.500000             1        1159.000000   \n",
       "1        NaN         0.250000             0        1828.666667   \n",
       "2        NaN         0.500000             6        1370.500000   \n",
       "3        NaN         0.933333             3        1742.666667   \n",
       "4        NaN         0.500000             1         611.500000   \n",
       "\n",
       "   manager_stability  BusinessTravel_Travel_Frequently  \\\n",
       "0           0.000000                             False   \n",
       "1           0.666667                             False   \n",
       "2           0.875000                             False   \n",
       "3           0.533333                             False   \n",
       "4           0.000000                              True   \n",
       "\n",
       "   BusinessTravel_Travel_Rarely  Department_Research & Development  \\\n",
       "0                         False                               True   \n",
       "1                          True                              False   \n",
       "2                          True                               True   \n",
       "3                          True                               True   \n",
       "4                         False                               True   \n",
       "\n",
       "   Department_Sales  EducationField_Life Sciences  EducationField_Marketing  \\\n",
       "0             False                         False                     False   \n",
       "1              True                         False                      True   \n",
       "2             False                          True                     False   \n",
       "3             False                          True                     False   \n",
       "4             False                          True                     False   \n",
       "\n",
       "   EducationField_Medical  EducationField_Other  \\\n",
       "0                    True                 False   \n",
       "1                   False                 False   \n",
       "2                   False                 False   \n",
       "3                   False                 False   \n",
       "4                   False                 False   \n",
       "\n",
       "   EducationField_Technical Degree  Gender_Male  JobRole_Human Resources  \\\n",
       "0                            False         True                    False   \n",
       "1                            False         True                    False   \n",
       "2                            False         True                    False   \n",
       "3                            False         True                    False   \n",
       "4                            False         True                    False   \n",
       "\n",
       "   JobRole_Laboratory Technician  JobRole_Manager  \\\n",
       "0                           True            False   \n",
       "1                          False            False   \n",
       "2                          False            False   \n",
       "3                          False            False   \n",
       "4                           True            False   \n",
       "\n",
       "   JobRole_Manufacturing Director  JobRole_Research Director  \\\n",
       "0                           False                      False   \n",
       "1                           False                      False   \n",
       "2                           False                      False   \n",
       "3                           False                      False   \n",
       "4                           False                      False   \n",
       "\n",
       "   JobRole_Research Scientist  JobRole_Sales Executive  \\\n",
       "0                       False                    False   \n",
       "1                       False                     True   \n",
       "2                        True                    False   \n",
       "3                       False                    False   \n",
       "4                       False                    False   \n",
       "\n",
       "   JobRole_Sales Representative  MaritalStatus_Married  MaritalStatus_Single  \n",
       "0                         False                  False                  True  \n",
       "1                         False                  False                 False  \n",
       "2                         False                  False                 False  \n",
       "3                         False                  False                  True  \n",
       "4                         False                  False                  True  "
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "f2ff65ab-72be-4836-8a47-1cb99323d5ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "x=train.drop(['Attrition'],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "698849e6-19ac-406e-b16d-f039d7bfe2c6",
   "metadata": {},
   "outputs": [],
   "source": [
    "y=train['Attrition']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "b833d216-0d4d-4a9d-b0f5-a77ea5b370f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train,x_val, y_train, y_val=train_test_split(x,y, test_size=.20, random_state=42, stratify=y)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e2a78fd-3c75-47b7-8099-6eb43349f409",
   "metadata": {},
   "source": [
    "### 1th LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "006fb62d-5011-4939-87e1-4a8cb49966d5",
   "metadata": {},
   "outputs": [],
   "source": [
    "lr=LogisticRegression(max_iter=1000,class_weight='balanced',n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "0c2ee3ae-d06c-4889-830c-532cd63251d3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: #000;\n",
       "  --sklearn-color-text-muted: #666;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\n",
       "}\n",
       "\n",
       "#sk-container-id-1 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-1 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-1 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: flex;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "  align-items: start;\n",
       "  justify-content: space-between;\n",
       "  gap: 0.5em;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 label.sk-toggleable__label .caption {\n",
       "  font-size: 0.6rem;\n",
       "  font-weight: lighter;\n",
       "  color: var(--sklearn-color-text-muted);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content {\n",
       "  display: none;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  overflow: visible;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-1 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-1 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-1 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  display: inline-block;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-1 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-1 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 0.5em;\n",
       "  text-align: center;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-1 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".estimator-table summary {\n",
       "    padding: .5rem;\n",
       "    font-family: monospace;\n",
       "    cursor: pointer;\n",
       "}\n",
       "\n",
       ".estimator-table details[open] {\n",
       "    padding-left: 0.1rem;\n",
       "    padding-right: 0.1rem;\n",
       "    padding-bottom: 0.3rem;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table {\n",
       "    margin-left: auto !important;\n",
       "    margin-right: auto !important;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:nth-child(odd) {\n",
       "    background-color: #fff;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:nth-child(even) {\n",
       "    background-color: #f6f6f6;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:hover {\n",
       "    background-color: #e0e0e0;\n",
       "}\n",
       "\n",
       ".estimator-table table td {\n",
       "    border: 1px solid rgba(106, 105, 104, 0.232);\n",
       "}\n",
       "\n",
       ".user-set td {\n",
       "    color:rgb(255, 94, 0);\n",
       "    text-align: left;\n",
       "}\n",
       "\n",
       ".user-set td.value pre {\n",
       "    color:rgb(255, 94, 0) !important;\n",
       "    background-color: transparent !important;\n",
       "}\n",
       "\n",
       ".default td {\n",
       "    color: black;\n",
       "    text-align: left;\n",
       "}\n",
       "\n",
       ".user-set td i,\n",
       ".default td i {\n",
       "    color: black;\n",
       "}\n",
       "\n",
       ".copy-paste-icon {\n",
       "    background-image: url(data:image/svg+xml;base64,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);\n",
       "    background-repeat: no-repeat;\n",
       "    background-size: 14px 14px;\n",
       "    background-position: 0;\n",
       "    display: inline-block;\n",
       "    width: 14px;\n",
       "    height: 14px;\n",
       "    cursor: pointer;\n",
       "}\n",
       "</style><body><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>LogisticRegression(class_weight=&#x27;balanced&#x27;, max_iter=1000, n_jobs=-1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>LogisticRegression</div></div><div><a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.7/modules/generated/sklearn.linear_model.LogisticRegression.html\">?<span>Documentation for LogisticRegression</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></div></label><div class=\"sk-toggleable__content fitted\" data-param-prefix=\"\">\n",
       "        <div class=\"estimator-table\">\n",
       "            <details>\n",
       "                <summary>Parameters</summary>\n",
       "                <table class=\"parameters-table\">\n",
       "                  <tbody>\n",
       "                    \n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('penalty',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">penalty&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;l2&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('dual',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">dual&nbsp;</td>\n",
       "            <td class=\"value\">False</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('tol',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">tol&nbsp;</td>\n",
       "            <td class=\"value\">0.0001</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('C',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">C&nbsp;</td>\n",
       "            <td class=\"value\">1.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('fit_intercept',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">fit_intercept&nbsp;</td>\n",
       "            <td class=\"value\">True</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('intercept_scaling',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">intercept_scaling&nbsp;</td>\n",
       "            <td class=\"value\">1</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('class_weight',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">class_weight&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;balanced&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('random_state',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">random_state&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('solver',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">solver&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;lbfgs&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_iter',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_iter&nbsp;</td>\n",
       "            <td class=\"value\">1000</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('multi_class',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">multi_class&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;deprecated&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('verbose',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">verbose&nbsp;</td>\n",
       "            <td class=\"value\">0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('warm_start',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">warm_start&nbsp;</td>\n",
       "            <td class=\"value\">False</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_jobs',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">n_jobs&nbsp;</td>\n",
       "            <td class=\"value\">-1</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('l1_ratio',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">l1_ratio&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "                  </tbody>\n",
       "                </table>\n",
       "            </details>\n",
       "        </div>\n",
       "    </div></div></div></div></div><script>function copyToClipboard(text, element) {\n",
       "    // Get the parameter prefix from the closest toggleable content\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${text}` : text;\n",
       "\n",
       "    const originalStyle = element.style;\n",
       "    const computedStyle = window.getComputedStyle(element);\n",
       "    const originalWidth = computedStyle.width;\n",
       "    const originalHTML = element.innerHTML.replace('Copied!', '');\n",
       "\n",
       "    navigator.clipboard.writeText(fullParamName)\n",
       "        .then(() => {\n",
       "            element.style.width = originalWidth;\n",
       "            element.style.color = 'green';\n",
       "            element.innerHTML = \"Copied!\";\n",
       "\n",
       "            setTimeout(() => {\n",
       "                element.innerHTML = originalHTML;\n",
       "                element.style = originalStyle;\n",
       "            }, 2000);\n",
       "        })\n",
       "        .catch(err => {\n",
       "            console.error('Failed to copy:', err);\n",
       "            element.style.color = 'red';\n",
       "            element.innerHTML = \"Failed!\";\n",
       "            setTimeout(() => {\n",
       "                element.innerHTML = originalHTML;\n",
       "                element.style = originalStyle;\n",
       "            }, 2000);\n",
       "        });\n",
       "    return false;\n",
       "}\n",
       "\n",
       "document.querySelectorAll('.fa-regular.fa-copy').forEach(function(element) {\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const paramName = element.parentElement.nextElementSibling.textContent.trim();\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${paramName}` : paramName;\n",
       "\n",
       "    element.setAttribute('title', fullParamName);\n",
       "});\n",
       "</script></body>"
      ],
      "text/plain": [
       "LogisticRegression(class_weight='balanced', max_iter=1000, n_jobs=-1)"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr.fit(x_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "116d38ad-04ad-4f5a-b3c2-51364139bf3d",
   "metadata": {},
   "outputs": [],
   "source": [
    "lrtahmin=lr.predict(x_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "e66cdea5-3af0-4b11-842e-5c7fe6faa10f",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_val_proba=lr.predict_proba(x_val)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "c3344ff3-016d-4ca2-9647-018746df43e6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC: 0.7062\n"
     ]
    }
   ],
   "source": [
    "roc_auc=roc_auc_score(y_val, y_val_proba)\n",
    "print(f\"ROC AUC: {roc_auc:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "6b27318e-87ce-4fa6-a7d6-a62495ea0799",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.6815\n"
     ]
    }
   ],
   "source": [
    "acc=accuracy_score(y_val, lrtahmin)\n",
    "print(f\"Accuracy: {acc:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "c191359d-f876-423d-83e3-b97ac20a96eb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[204,  92],\n",
       "       [ 15,  25]])"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(y_val, lrtahmin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "91b3155e-3093-4683-aa33-87c72b722c0c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.93      0.69      0.79       296\n",
      "         1.0       0.21      0.62      0.32        40\n",
      "\n",
      "    accuracy                           0.68       336\n",
      "   macro avg       0.57      0.66      0.56       336\n",
      "weighted avg       0.85      0.68      0.74       336\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"\\nClassification Report:\")\n",
    "print(classification_report(y_val, lrtahmin))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "a1649fbb-d246-4099-bfbf-74a16c37db4e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr, tpr, thresholds=roc_curve(y_val, y_val_proba)\n",
    "roc_auc=auc(fpr, tpr)\n",
    "\n",
    "plt.figure(figsize=(6,5))\n",
    "plt.plot(fpr, tpr, label=f'Logistic Regression (AUC = {roc_auc:.3f})')\n",
    "plt.plot([0, 1], [0, 1], 'k--', label='Random Classifier')\n",
    "\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curve')\n",
    "plt.legend(loc='lower right')\n",
    "plt.grid(alpha=0.3)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "b6205de0-7b06-4be6-90f2-6f51e7d25836",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.naive_bayes import BernoulliNB\n",
    "\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score\n",
    "from sklearn.metrics import f1_score, confusion_matrix, classification_report\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "b = BernoulliNB()\n",
    "l = LogisticRegression()\n",
    "d = DecisionTreeClassifier()\n",
    "r = RandomForestClassifier()\n",
    "gb= GradientBoostingClassifier()\n",
    "kn= KNeighborsClassifier()\n",
    "ab= AdaBoostClassifier()\n",
    "mn= MultinomialNB()\n",
    "\n",
    "def algo_test(x, y):\n",
    "    modeller=[ b, l, d, r, gb, kn, ab, mn]\n",
    "    isimler=[\"BernoulliNB\", \"LogisticRegression\", \"DecisionTreeClassifier\", \n",
    "             \"RandomForestClassifier\", \"GradientBoostingClassifier\", \"KNeighborsClassifier\",\n",
    "             \"AdaBoostClassifier\", \"MultinomialNB\"]\n",
    "\n",
    "    x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=.3, random_state = 42)\n",
    "    \n",
    "    accuracy = []\n",
    "    precision = []\n",
    "    recall = []\n",
    "    f1 = []\n",
    "    mdl=[]\n",
    "\n",
    "    print(\"Veriler hazır modeller deneniyor\")\n",
    "    for model in modeller:\n",
    "        print(model, \" modeli eğitiliyor!..\")\n",
    "        model=model.fit(x_train,y_train)\n",
    "        tahmin=model.predict(x_test)\n",
    "        mdl.append(model)\n",
    "        accuracy.append(accuracy_score(y_test, tahmin))\n",
    "        precision.append(precision_score(y_test, tahmin, average=\"micro\"))\n",
    "        recall.append(recall_score(y_test, tahmin, average=\"micro\"))\n",
    "        f1.append(f1_score(y_test, tahmin, average=\"micro\"))\n",
    "        print(confusion_matrix(y_test, tahmin))\n",
    "\n",
    "    print(\"Eğitim tamamlandı.\")\n",
    "    \n",
    "    metrics=pd.DataFrame(columns=[\"Accuracy\", \"Precision\", \"Recall\", \"F1\", \"Model\"], index=isimler)\n",
    "    metrics[\"Accuracy\"] = accuracy\n",
    "    metrics[\"Precision\"] = precision  \n",
    "    metrics[\"Recall\"] = recall\n",
    "    metrics[\"F1\"] = f1\n",
    "    metrics[\"Model\"]=mdl\n",
    "\n",
    "    metrics.sort_values(\"F1\", ascending=False, inplace=True)\n",
    "\n",
    "    print(\"En başarılı model: \", metrics.iloc[0].name)\n",
    "    model=metrics.iloc[0,-1]\n",
    "    tahmin=model.predict(np.array(x_test) if model==kn else x_test)\n",
    "    print(\"Confusion Matrix:\")\n",
    "    print(confusion_matrix(y_test, tahmin))\n",
    "    print(\"classification Report:\")\n",
    "    print(classification_report(y_test, tahmin))\n",
    "    print(\"Diğer Modeller:\")\n",
    "    \n",
    "    return metrics.drop(\"Model\", axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "cda4bff1-df3f-4e3a-b3b3-9e5e94482f6c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Veriler hazır modeller deneniyor\n",
      "BernoulliNB()  modeli eğitiliyor!..\n",
      "[[403  36]\n",
      " [ 47  18]]\n",
      "LogisticRegression()  modeli eğitiliyor!..\n",
      "[[439   0]\n",
      " [ 65   0]]\n",
      "DecisionTreeClassifier()  modeli eğitiliyor!..\n",
      "[[394  45]\n",
      " [ 51  14]]\n",
      "RandomForestClassifier()  modeli eğitiliyor!..\n",
      "[[437   2]\n",
      " [ 58   7]]\n",
      "GradientBoostingClassifier()  modeli eğitiliyor!..\n",
      "[[430   9]\n",
      " [ 53  12]]\n",
      "KNeighborsClassifier()  modeli eğitiliyor!..\n",
      "[[433   6]\n",
      " [ 64   1]]\n",
      "AdaBoostClassifier()  modeli eğitiliyor!..\n",
      "[[430   9]\n",
      " [ 50  15]]\n",
      "MultinomialNB()  modeli eğitiliyor!..\n",
      "[[223 216]\n",
      " [ 27  38]]\n",
      "Eğitim tamamlandı.\n",
      "En başarılı model:  AdaBoostClassifier\n",
      "Confusion Matrix:\n",
      "[[430   9]\n",
      " [ 50  15]]\n",
      "classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.90      0.98      0.94       439\n",
      "         1.0       0.62      0.23      0.34        65\n",
      "\n",
      "    accuracy                           0.88       504\n",
      "   macro avg       0.76      0.61      0.64       504\n",
      "weighted avg       0.86      0.88      0.86       504\n",
      "\n",
      "Diğer Modeller:\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  File \"C:\\Users\\enesy\\anaconda3\\Lib\\site-packages\\joblib\\externals\\loky\\backend\\context.py\", line 257, in _count_physical_cores\n",
      "    cpu_info = subprocess.run(\n",
      "        \"wmic CPU Get NumberOfCores /Format:csv\".split(),\n",
      "        capture_output=True,\n",
      "        text=True,\n",
      "    )\n",
      "  File \"C:\\Users\\enesy\\anaconda3\\Lib\\subprocess.py\", line 554, in run\n",
      "    with Popen(*popenargs, **kwargs) as process:\n",
      "         ~~~~~^^^^^^^^^^^^^^^^^^^^^^\n",
      "  File \"C:\\Users\\enesy\\anaconda3\\Lib\\subprocess.py\", line 1039, in __init__\n",
      "    self._execute_child(args, executable, preexec_fn, close_fds,\n",
      "    ~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n",
      "                        pass_fds, cwd, env,\n",
      "                        ^^^^^^^^^^^^^^^^^^^\n",
      "    ...<5 lines>...\n",
      "                        gid, gids, uid, umask,\n",
      "                        ^^^^^^^^^^^^^^^^^^^^^^\n",
      "                        start_new_session, process_group)\n",
      "                        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n",
      "  File \"C:\\Users\\enesy\\anaconda3\\Lib\\subprocess.py\", line 1554, in _execute_child\n",
      "    hp, ht, pid, tid = _winapi.CreateProcess(executable, args,\n",
      "                       ~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^\n",
      "                             # no special security\n",
      "                             ^^^^^^^^^^^^^^^^^^^^^\n",
      "    ...<4 lines>...\n",
      "                             cwd,\n",
      "                             ^^^^\n",
      "                             startupinfo)\n",
      "                             ^^^^^^^^^^^^\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>Precision</th>\n",
       "      <th>Recall</th>\n",
       "      <th>F1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AdaBoostClassifier</th>\n",
       "      <td>0.882937</td>\n",
       "      <td>0.882937</td>\n",
       "      <td>0.882937</td>\n",
       "      <td>0.882937</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RandomForestClassifier</th>\n",
       "      <td>0.880952</td>\n",
       "      <td>0.880952</td>\n",
       "      <td>0.880952</td>\n",
       "      <td>0.880952</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GradientBoostingClassifier</th>\n",
       "      <td>0.876984</td>\n",
       "      <td>0.876984</td>\n",
       "      <td>0.876984</td>\n",
       "      <td>0.876984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LogisticRegression</th>\n",
       "      <td>0.871032</td>\n",
       "      <td>0.871032</td>\n",
       "      <td>0.871032</td>\n",
       "      <td>0.871032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KNeighborsClassifier</th>\n",
       "      <td>0.861111</td>\n",
       "      <td>0.861111</td>\n",
       "      <td>0.861111</td>\n",
       "      <td>0.861111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BernoulliNB</th>\n",
       "      <td>0.835317</td>\n",
       "      <td>0.835317</td>\n",
       "      <td>0.835317</td>\n",
       "      <td>0.835317</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DecisionTreeClassifier</th>\n",
       "      <td>0.809524</td>\n",
       "      <td>0.809524</td>\n",
       "      <td>0.809524</td>\n",
       "      <td>0.809524</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MultinomialNB</th>\n",
       "      <td>0.517857</td>\n",
       "      <td>0.517857</td>\n",
       "      <td>0.517857</td>\n",
       "      <td>0.517857</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            Accuracy  Precision    Recall        F1\n",
       "AdaBoostClassifier          0.882937   0.882937  0.882937  0.882937\n",
       "RandomForestClassifier      0.880952   0.880952  0.880952  0.880952\n",
       "GradientBoostingClassifier  0.876984   0.876984  0.876984  0.876984\n",
       "LogisticRegression          0.871032   0.871032  0.871032  0.871032\n",
       "KNeighborsClassifier        0.861111   0.861111  0.861111  0.861111\n",
       "BernoulliNB                 0.835317   0.835317  0.835317  0.835317\n",
       "DecisionTreeClassifier      0.809524   0.809524  0.809524  0.809524\n",
       "MultinomialNB               0.517857   0.517857  0.517857  0.517857"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "algo_test(x,y)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e71fcdd-ef4d-4049-8cbd-201bd349c46a",
   "metadata": {},
   "source": [
    "### 2th AdaBoostClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "7167a974-8b80-4adc-b9c1-bd9e5ff81b01",
   "metadata": {},
   "outputs": [],
   "source": [
    "ab=AdaBoostClassifier(estimator=DecisionTreeClassifier(max_depth=2, random_state=42),n_estimators=400,learning_rate=0.05,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "a7657429-99b8-40b7-9574-ec5ffab1dfef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: #000;\n",
       "  --sklearn-color-text-muted: #666;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\n",
       "}\n",
       "\n",
       "#sk-container-id-2 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-2 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-2 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: flex;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "  align-items: start;\n",
       "  justify-content: space-between;\n",
       "  gap: 0.5em;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 label.sk-toggleable__label .caption {\n",
       "  font-size: 0.6rem;\n",
       "  font-weight: lighter;\n",
       "  color: var(--sklearn-color-text-muted);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable__content {\n",
       "  display: none;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  overflow: visible;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-2 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-2 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-2 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  display: inline-block;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-2 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-2 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 0.5em;\n",
       "  text-align: center;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-2 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-2 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".estimator-table summary {\n",
       "    padding: .5rem;\n",
       "    font-family: monospace;\n",
       "    cursor: pointer;\n",
       "}\n",
       "\n",
       ".estimator-table details[open] {\n",
       "    padding-left: 0.1rem;\n",
       "    padding-right: 0.1rem;\n",
       "    padding-bottom: 0.3rem;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table {\n",
       "    margin-left: auto !important;\n",
       "    margin-right: auto !important;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:nth-child(odd) {\n",
       "    background-color: #fff;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:nth-child(even) {\n",
       "    background-color: #f6f6f6;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:hover {\n",
       "    background-color: #e0e0e0;\n",
       "}\n",
       "\n",
       ".estimator-table table td {\n",
       "    border: 1px solid rgba(106, 105, 104, 0.232);\n",
       "}\n",
       "\n",
       ".user-set td {\n",
       "    color:rgb(255, 94, 0);\n",
       "    text-align: left;\n",
       "}\n",
       "\n",
       ".user-set td.value pre {\n",
       "    color:rgb(255, 94, 0) !important;\n",
       "    background-color: transparent !important;\n",
       "}\n",
       "\n",
       ".default td {\n",
       "    color: black;\n",
       "    text-align: left;\n",
       "}\n",
       "\n",
       ".user-set td i,\n",
       ".default td i {\n",
       "    color: black;\n",
       "}\n",
       "\n",
       ".copy-paste-icon {\n",
       "    background-image: url(data:image/svg+xml;base64,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);\n",
       "    background-repeat: no-repeat;\n",
       "    background-size: 14px 14px;\n",
       "    background-position: 0;\n",
       "    display: inline-block;\n",
       "    width: 14px;\n",
       "    height: 14px;\n",
       "    cursor: pointer;\n",
       "}\n",
       "</style><body><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>AdaBoostClassifier(estimator=DecisionTreeClassifier(max_depth=2,\n",
       "                                                    random_state=42),\n",
       "                   learning_rate=0.05, n_estimators=400, random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>AdaBoostClassifier</div></div><div><a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.7/modules/generated/sklearn.ensemble.AdaBoostClassifier.html\">?<span>Documentation for AdaBoostClassifier</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></div></label><div class=\"sk-toggleable__content fitted\" data-param-prefix=\"\">\n",
       "        <div class=\"estimator-table\">\n",
       "            <details>\n",
       "                <summary>Parameters</summary>\n",
       "                <table class=\"parameters-table\">\n",
       "                  <tbody>\n",
       "                    \n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('estimator',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">estimator&nbsp;</td>\n",
       "            <td class=\"value\">DecisionTreeC...ndom_state=42)</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_estimators',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">n_estimators&nbsp;</td>\n",
       "            <td class=\"value\">400</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('learning_rate',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">learning_rate&nbsp;</td>\n",
       "            <td class=\"value\">0.05</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('algorithm',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">algorithm&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;deprecated&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('random_state',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">random_state&nbsp;</td>\n",
       "            <td class=\"value\">42</td>\n",
       "        </tr>\n",
       "    \n",
       "                  </tbody>\n",
       "                </table>\n",
       "            </details>\n",
       "        </div>\n",
       "    </div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>estimator: DecisionTreeClassifier</div></div></label><div class=\"sk-toggleable__content fitted\" data-param-prefix=\"estimator__\"><pre>DecisionTreeClassifier(max_depth=2, random_state=42)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>DecisionTreeClassifier</div></div><div><a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.7/modules/generated/sklearn.tree.DecisionTreeClassifier.html\">?<span>Documentation for DecisionTreeClassifier</span></a></div></label><div class=\"sk-toggleable__content fitted\" data-param-prefix=\"estimator__\">\n",
       "        <div class=\"estimator-table\">\n",
       "            <details>\n",
       "                <summary>Parameters</summary>\n",
       "                <table class=\"parameters-table\">\n",
       "                  <tbody>\n",
       "                    \n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('criterion',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">criterion&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;gini&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('splitter',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">splitter&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;best&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_depth',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_depth&nbsp;</td>\n",
       "            <td class=\"value\">2</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_samples_split',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_samples_split&nbsp;</td>\n",
       "            <td class=\"value\">2</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_samples_leaf',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_samples_leaf&nbsp;</td>\n",
       "            <td class=\"value\">1</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_weight_fraction_leaf',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_weight_fraction_leaf&nbsp;</td>\n",
       "            <td class=\"value\">0.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_features',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_features&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('random_state',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">random_state&nbsp;</td>\n",
       "            <td class=\"value\">42</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_leaf_nodes',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_leaf_nodes&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_impurity_decrease',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_impurity_decrease&nbsp;</td>\n",
       "            <td class=\"value\">0.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('class_weight',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">class_weight&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('ccp_alpha',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">ccp_alpha&nbsp;</td>\n",
       "            <td class=\"value\">0.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('monotonic_cst',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">monotonic_cst&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "                  </tbody>\n",
       "                </table>\n",
       "            </details>\n",
       "        </div>\n",
       "    </div></div></div></div></div></div></div></div></div></div><script>function copyToClipboard(text, element) {\n",
       "    // Get the parameter prefix from the closest toggleable content\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${text}` : text;\n",
       "\n",
       "    const originalStyle = element.style;\n",
       "    const computedStyle = window.getComputedStyle(element);\n",
       "    const originalWidth = computedStyle.width;\n",
       "    const originalHTML = element.innerHTML.replace('Copied!', '');\n",
       "\n",
       "    navigator.clipboard.writeText(fullParamName)\n",
       "        .then(() => {\n",
       "            element.style.width = originalWidth;\n",
       "            element.style.color = 'green';\n",
       "            element.innerHTML = \"Copied!\";\n",
       "\n",
       "            setTimeout(() => {\n",
       "                element.innerHTML = originalHTML;\n",
       "                element.style = originalStyle;\n",
       "            }, 2000);\n",
       "        })\n",
       "        .catch(err => {\n",
       "            console.error('Failed to copy:', err);\n",
       "            element.style.color = 'red';\n",
       "            element.innerHTML = \"Failed!\";\n",
       "            setTimeout(() => {\n",
       "                element.innerHTML = originalHTML;\n",
       "                element.style = originalStyle;\n",
       "            }, 2000);\n",
       "        });\n",
       "    return false;\n",
       "}\n",
       "\n",
       "document.querySelectorAll('.fa-regular.fa-copy').forEach(function(element) {\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const paramName = element.parentElement.nextElementSibling.textContent.trim();\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${paramName}` : paramName;\n",
       "\n",
       "    element.setAttribute('title', fullParamName);\n",
       "});\n",
       "</script></body>"
      ],
      "text/plain": [
       "AdaBoostClassifier(estimator=DecisionTreeClassifier(max_depth=2,\n",
       "                                                    random_state=42),\n",
       "                   learning_rate=0.05, n_estimators=400, random_state=42)"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ab.fit(x_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "31025893-5620-4e65-8c92-9c4416020a1a",
   "metadata": {},
   "outputs": [],
   "source": [
    "abtahmin=ab.predict(x_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "663d4be7-0f55-4752-bc0e-3cf6f71b26ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_proba_ab=ab.predict_proba(x_val)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "5671308f-7fc5-40c9-839e-a1d7512d7090",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC: 0.7834\n"
     ]
    }
   ],
   "source": [
    "roc_auc=roc_auc_score(y_val, y_proba_ab)\n",
    "print(f\"ROC AUC: {roc_auc:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "69d5ac4d-1dc9-4e3d-b499-a76eda07029c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.8929\n"
     ]
    }
   ],
   "source": [
    "acc=accuracy_score(y_val, abtahmin)\n",
    "print(f\"Accuracy: {acc:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "b851f76f-5aee-44f1-ac52-c316a089a7c8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[293,   3],\n",
       "       [ 33,   7]])"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(y_val, abtahmin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "4db1e288-1bd6-4287-b907-6b52c3f1ca75",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.90      0.99      0.94       296\n",
      "         1.0       0.70      0.17      0.28        40\n",
      "\n",
      "    accuracy                           0.89       336\n",
      "   macro avg       0.80      0.58      0.61       336\n",
      "weighted avg       0.88      0.89      0.86       336\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"\\nClassification Report:\")\n",
    "print(classification_report(y_val, abtahmin))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "82b72cf7-7592-4ea0-a2a6-e9ccedc9e799",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhgAAAHUCAYAAAB4RlFCAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAfEpJREFUeJzt3XdYFNf7NvB7l94tFEURsPdoxB41GiOWiF2MvWA3FqLG6NeuMZYYNRGMDaKxYOwaG7H3ghqNGDWKYsEgioD03T3vH/7Y15UiCwMD7P25rr0u9syZmWfPDrvPnjlzRiGEECAiIiKSkFLuAIiIiKjoYYJBREREkmOCQURERJJjgkFERESSY4JBREREkmOCQURERJJjgkFERESSY4JBREREkmOCQURERJJjgkFkAAIDA6FQKLQPY2NjlC5dGr169cK9e/cyXCc1NRX+/v5o3Lgx7OzsYGFhgWrVqmHKlCl4+fJlhutoNBps3LgRrVu3hr29PUxMTODo6IgvvvgC+/btg0aj+WCsycnJ+Pnnn/HJJ5+gePHiMDU1RZkyZdCzZ0+cPHkyV+1ARPmHCQaRAQkICMD58+fx559/YsyYMdi7dy8++eQTREdH69RLSEjA559/jq+++gp169bFli1bcODAAfTr1w+rV69G3bp1cefOHZ11kpKS0L59ewwYMACOjo7w9/fHsWPHsGrVKjg7O6NHjx7Yt29flvFFRUWhadOm8PX1Rc2aNREYGIijR4/ihx9+gJGRET777DP89ddfkrcLEeUBQURFXkBAgAAgLl++rFM+e/ZsAUCsX79ep3zYsGECgNi6dWu6bd25c0fY2dmJGjVqCJVKpS0fOXKkACB+/fXXDGO4e/eu+Ouvv7KMs127dsLY2FgcPXo0w+WXLl0Sjx49ynIb2ZWQkCDJdogoY+zBIDJgHh4eAID//vtPW/b8+XOsX78enp6e8Pb2TrdO5cqV8c033+DWrVvYvXu3dp21a9fC09MT/fv3z3BflSpVQu3atTONJSQkBAcPHsSQIUPQqlWrDOvUr18f5cqVAwDMmjULCoUiXZ2000EPHz7Ulrm5ueGLL77Azp07UbduXZibm2P27NmoW7cumjVrlm4barUaZcqUQdeuXbVlKSkpmDdvHqpWrQozMzM4ODhg0KBBePHiRaaviciQMcEgMmBhYWEA3iYNaY4fPw6VSoXOnTtnul7asuDgYO06qampWa7zIUeOHNHZttSuXr2KSZMmYezYsTh06BC6deuGQYMG4cyZM+nGoRw5cgTPnj3DoEGDALwdW9KpUyd8//336N27N/744w98//33CA4OxqefforExMQ8iZmoMDOWOwAiyj9qtRoqlQpJSUk4e/Ys5s2bh+bNm8PLy0tbJzw8HADg7u6e6XbSlqXVzc46HyLFNrISGRmJ0NBQnWSqfPnymDRpEgIDAzF//nxteWBgIJycnNCuXTsAwLZt23Do0CHs2LFDp1fjo48+Qv369REYGIiRI0fmSdxEhRV7MIgMSKNGjWBiYgIbGxu0bdsWxYsXx549e2BsnLPfGhmdoiioateurZNcAEDJkiXRsWNH/Prrr9orXKKjo7Fnzx70799f2y779+9HsWLF0LFjR6hUKu2jTp06KFWqFE6cOJHfL4eowGOCQWRANmzYgMuXL+PYsWMYPnw4bt++jS+//FKnTtoYh7TTJxlJW+bi4pLtdT5Eim1kpXTp0hmWDx48GE+fPtWe7tmyZQuSk5MxcOBAbZ3//vsPr1+/hqmpKUxMTHQez58/R1RUVJ7ETFSYMcEgMiDVqlWDh4cHWrZsiVWrVsHHxweHDh3C9u3btXVatmwJY2Nj7QDOjKQt+/zzz7XrmJiYZLnOh3h6eups+0PMzc0BvJ03412Zfdln1tvi6ekJZ2dnBAQEAHh7KW/Dhg1RvXp1bR17e3uULFkSly9fzvDh5+eXrZiJDIrcl7EQUd7L7DLVV69eieLFi4tq1aoJtVqtLc+Ly1T//fffXF+mevnyZe1lqlu2bBEAxKVLl3TqNG/eXAAQYWFh2jJXV1fRoUOHTPf7zTffCDMzM3Hq1CkBQPzyyy86y3/77TcBQFy4cCHL+Ino/2OCQWQAMkswhBBi0aJFAoDYuHGjtuzNmzeiRYsWwtjYWIwaNUocPHhQHDt2THz33XeiRIkSomzZsuKff/7R2U5iYqLw9PQUCoVC9O7dW/z+++/i1KlTYufOnWLkyJHC3Nxc7N69O8s4X7x4IerVqydMTU3FiBEjxJ49e8SpU6dEUFCQ6Nu3rzAyMhLXr18XQggRExMjSpQoIWrVqiV27dol9u3bJ7p16ybc3d31TjDu3LkjAIiyZcsKCwsL8fr1a53lKpVKtGvXTpQoUULMnj1bHDx4UPz5558iMDBQDBgwQOzcuTPL10VkiJhgEBmArBKMxMREUa5cOVGpUiWdHomUlBSxcuVK0bBhQ2FtbS3MzMxElSpVxOTJk0VUVFSG+1GpVOLXX38VrVq1EiVKlBDGxsbCwcFBtGvXTmzevFmnlyQziYmJYsWKFaJx48bC1tZWGBsbC2dnZ9G1a1fxxx9/6NS9dOmSaNKkibCyshJlypQRM2fOFGvXrtU7wRBCiCZNmggAok+fPhkuT01NFUuWLBEfffSRMDc3F9bW1qJq1api+PDh4t69ex98XUSGRiGEEHKcmiEiIqKii4M8iYiISHJMMIiIiEhyTDCIiIhIckwwiIiISHJMMIiIiEhyTDCIiIhIcgZ3N1WNRoNnz57BxsamUN2oiYiISG5CCMTFxcHZ2RlKZdZ9FAaXYDx79kx7gyYiIiLS3+PHj1G2bNks6xhcgmFjYwPgbePY2tpKtl2NRoMXL17AwcHhg1kdfRjbU3psU2mxPaXHNpVWXrRnbGwsXFxctN+lWTG4BCPttIitra3kCUZSUhJsbW35jyEBtqf02KbSYntKj20qrbxsz+wMMeA7SERERJJjgkFERESSY4JBREREkmOCQURERJJjgkFERESSY4JBREREkmOCQURERJJjgkFERESSY4JBREREkmOCQURERJKTNcE4deoUOnbsCGdnZygUCuzevfuD65w8eRL16tWDubk5ypcvj1WrVuV9oERERKQXWROM+Ph4fPTRR/j555+zVT8sLAzt27dHs2bNcO3aNUydOhVjx47Fjh078jhSIiIi0oesNztr164d2rVrl+36q1atQrly5bBs2TIAQLVq1XDlyhUsWbIE3bp1y6MoiYiosBBCIDFVLXcYBYJGo0FiqhpCCFn2X6jupnr+/Hm0adNGp8zT0xPr1q1DamoqTExM0q2TnJyM5ORk7fPY2FgAbxteo9FIFptGo4EQQtJtGjK2p/TYptJie0ovt20qhEDPXy4gJPy1tIEVQkJokPggBJYV6uPGjM9gbf7hu59mhz7vTaFKMJ4/fw4nJyedMicnJ6hUKkRFRaF06dLp1lmwYAFmz56drvzFixdISkqSLDaNRoOYmBgIIXibYQmwPaXHNpUW21N6uW3TxFQ1kwsAmpQkvDzwIxLunEWJNqPw4kVtJJil/wGeE3FxcdmuW6gSDCD9PejTun4yuzf9t99+C19fX+3z2NhYuLi4wMHBAba2tpLFpdFooFAo4ODgwA8bCbA9pcc2lRbbU3q5bdOEFJX270tTW8HS1EjK8AqFJ48fo1fP7nh85zpMTEzwzeflUc65FIyMpGkLc3PzbNctVAlGqVKl8Pz5c52yyMhIGBsbo2TJkhmuY2ZmBjMzs3TlSqVS8g8FhUKRJ9s1VGxP6bFNpcX2lF5u2vTddazNTWBpWqi+4nLtwoUL6Ny5M/777z84ODhg+/btqFy5MoyMjCQ7RvXZTqH6r2jcuDGCg4N1yo4cOQIPD48Mx18QEREZgo0bN+LTTz/Ff//9h9q1a+Py5cv45JNPZI1J1gTjzZs3uH79Oq5fvw7g7WWo169fR3h4OIC3pzf69++vrT9ixAg8evQIvr6+uH37NtavX49169Zh4sSJcoRPREQku3/++QcDBgxAcnIyOnfujLNnz8LV1VXusOQ9RXLlyhW0bNlS+zxtrMSAAQMQGBiIiIgIbbIBAO7u7jhw4AAmTJiAlStXwtnZGStWrOAlqkREZLCqVq2KhQsXIiYmBnPmzCkwp+xkTTA+/fTTLK/PDQwMTFfWokULXL16NQ+jIiIiKtjCwsKgUCjg5uYGAJg0aZK8AWXAsEbAEBGRZKSe1CptYqiEFFUOryIxjAm2Tp06ha5du6J06dI4d+4cbGxs5A4pQ0wwiIhIb0IIdF91HiGPouUOxaCsWbMGo0aNgkqlgpubG+Lj4wtsglEwTtQQEVGhkpiqLrDJhYdrcViYFK05MFQqFcaNG4dhw4ZBpVLB29sbp06dQqlSpeQOLVPswSAioly58r/WkkxqpdFo8OLFi1xPXmZhYpTp5IuFUXR0NLy9vbXTNMydOxfTpk0r8K+RCQYREeWKpamRJJNaaTQaWJi83VZBuRKiIBg9ejSCg4NhaWmJjRs3omvXrnKHlC1MMIiIiAqwJUuWICwsDP7+/qhTp47c4WQbU0QiIqICRAiBS5cuaZ87Ozvj3LlzhSq5AJhgEBERFRipqakYOXIkGjZsiKCgIG15QR9vkRGeIiEiIioAXr58ie7du+PEiRNQKBTpbu5Z2DDBICIycDmZMMtQJrXKL7du3YKXlxcePHgAGxsbbNmyBR06dJA7rFxhgkFEZMA4YZb89u/fj969eyMuLg7ly5fH3r17UaNGDbnDyjUmGEREBiy3E2YVxUmt8tPt27fRqVMnaDQatGjRAtu3b4e9vb3cYUmCCQYREQHI2YRZRW1Sq/xWrVo1fP3114iJicFPP/0EU1NTuUOSDBMMIiICIN2EWZS158+fQ6lUwtHREQDw/fffQ6FQFLlEjZepEhER5ZNr166hQYMG6Nq1K5KTkwEASqWyyCUXABMMIiKifLFjxw588sknePz4MV68eIHIyEi5Q8pTTDCIiIjykBACc+fORffu3ZGQkIA2bdrg4sWLcHFxkTu0PMWTbURE+SAnc03kB85nkbcSEhIwePBg7ayc48aNw5IlS2BsXPS/fov+KyQikhnnmjBcQ4cORVBQEExMTODn5wcfHx+5Q8o3PEVCRJTHcjvXRH7gfBZ5Y9asWahcuTL+/PNPg0ouAPZgEBHlq5zMNZEfOJ+FdP79919UrFgRAFCpUiWEhobCyKjgved5jQkGEVE+4lwTRZdGo8H//vc/LFq0CAcPHsTnn38OAAaZXAA8RUJERJRrcXFx6Nq1KxYsWAC1Wo2LFy/KHZLsmEYTERHlwsOHD+Hl5YWbN2/CzMwMa9euRd++feUOS3ZMMIiIiHLo9OnT6Nq1K6KiouDk5ITdu3ejUaNGcodVIDDBICIiyoG///4bn332GVJTU1G3bl3s2bOnyE+epQ8mGERE78nOpFgajQaJqWokpKigVGY9nI2TWRVNNWrUQN++ffHmzRsEBATAyspK7pAKFCYYRETv4KRYlJWYmBgoFArY2tpCoVDgl19+gZGR0QeTTEPEFiEiekdeTorFyawKt3///ReNGjVC7969oVa/7ZUyMTFhcpEJ9mAQEWUiq0mxNBoNXrx4AQcHh2x/wXAyq8Lr2LFj6N69O6Kjo/HmzRs8efIErq6ucodVoDHBICLKRFaTYmk0GliYvF3OX7BFm5+fH8aOHQu1Wo2GDRti165dKF26tNxhFXj8ryAiIspAamoqRo0ahdGjR0OtVqNv3744ceIEk4tsYoJBRESUgSFDhsDf3x8KhQLff/89NmzYAHNzc7nDKjSYYBAREWVg3LhxcHR0xO7du/HNN99w/IyeOAaDiIjo/0RGRsLR0REAUK9ePYSFhcHS0lLmqAon9mAQEeHt/BcJKSpOimWghBBYunQp3N3dcenSJW05k4ucYw8GERk8Tq5l2JKTkzFy5EgEBAQAAHbt2oUGDRrIHFXhxwSDiAxeRpNrcVIswxAZGYmuXbvi7NmzUCqVWLp0KcaOHSt3WEUCEwwionekTa7FSbGKvhs3bqBjx44IDw+HnZ0dgoKC4OnpKXdYRQYTDCKid2Q1uRYVHX///TeaNGmC+Ph4VKpUCfv27UOVKlXkDqtI4X8REREZnOrVq6N169aIj4/Htm3bULx4cblDKnKYYBARkUFITEyEQqGAubk5lEolNm/eDFNTUxgb86swL/AyVSIiKvKePXuG5s2bY9iwYRBCAHh7CSqTi7zDliUioiLt8uXL6Ny5M549e4YHDx4gPDycd0LNB+zBIKIi7/9PopXZg5NrFVVbt25F8+bN8ezZM1SvXh2XLl1icpFP2INBREUaJ9EyTBqNBjNnzsS8efMAAO3bt8eWLVtga2src2SGgz0YRFSkZTSJVmY4uVbRMXz4cG1yMXHiROzdu5fJRT5jDwYRGYy0SbQyw8m1io7evXtjy5YtWLlyJQYMGCB3OAaJCQYRGQxOolW0xcfHw8rKCgDQsmVLPHz4EPb29jJHZbh4ioSIiAq9X3/9FeXLl8c///yjLWNyIS8mGEREVGip1WpMmjQJAwcORGRkJH755Re5Q6L/w75CIiIqlGJjY/Hll1/iwIEDAIDp06dj1qxZ8gZFWkwwiEhLCIHE1KI1JwTnuCia7t+/Dy8vL4SGhsLc3ByBgYHw9vaWOyx6BxMMIgLA+SKo8Lh16xaaN2+OV69ewdnZGXv27IGHh4fcYdF7mGAQEQD95osojDjHRdFRqVIl1KxZE4mJidi9ezecnZ3lDokywASDiNL50HwRhRHnuCjcVCoVAMDY2BimpqbYtWsXLCwsYGFhIXNklBkmGESUDueLoIIkOjoaPXv2RK1atbB06VIAQIkSJWSOij6El6kSEVGB9c8//6Bhw4b4888/sXr1aoSHh8sdEmUTEwwiIiqQDh8+jEaNGuHevXtwdXXFuXPnUK5cObnDomySPcHw8/ODu7s7zM3NUa9ePZw+fTrL+ps2bcJHH30ES0tLlC5dGoMGDcLLly/zKVoiIsprQggsX74c7du3R0xMDD755BNcunQJtWvXljs00oOsCUZQUBDGjx+PadOm4dq1a2jWrBnatWuXaRfYmTNn0L9/fwwZMgS3bt3C77//jsuXL8PHxyefIyciorzi6+uL8ePHQ6PRYNCgQfjzzz/h6Ogod1ikJ1kTjKVLl2LIkCHw8fFBtWrVsGzZMri4uMDf3z/D+hcuXICbmxvGjh0Ld3d3fPLJJxg+fDiuXLmSz5ETFR5CCCSkqJCQokJiqlr7d/oHJ6SigqFZs2YwMjLC0qVLsW7dOpiZmckdEuWAbMPEU1JSEBISgilTpuiUt2nTBufOnctwnSZNmmDatGk4cOAA2rVrh8jISGzfvh0dOnTIdD/JyclITk7WPo+NjQUAaDQaaDQaCV4JtNsTQki6TUPG9pSGEAI9f7mAkPDXeq0n9f9HUcRjVFoqlQpKpRJCCHTu3Bl37tyBu7s7hBAQQsgdXqGUF8eoPtuSLcGIioqCWq2Gk5OTTrmTkxOeP3+e4TpNmjTBpk2b4O3tjaSkJKhUKnh5eeGnn37KdD8LFizA7Nmz05W/ePECSUlJuXsR79BoNIiJiYEQAkql7ENbCj22pzQSU9V6Jxe1na0QF/0SbzhnRJZ4jErn8OHDmD17NoKCgmBtbQ0hBKysrBAZGSl3aIVaXhyjcXFx2a4r+4Xu7098I4TIdDKc0NBQjB07FjNmzICnpyciIiIwadIkjBgxAuvWrctwnW+//Ra+vr7a57GxsXBxcYGDgwNsbW0lex0ajQYKhQIODg78sJEA21MaCSkq7d8Xp3yK+Nho2NvbZ9mmnJAqe3iM5p4QAosWLcK0adMghEBgYCD+97//sU0lkhfHqLm5ebbrypZg2Nvbw8jIKF1vRWRkZLpejTQLFixA06ZNMWnSJABA7dq1YWVlhWbNmmHevHkoXbp0unXMzMwyPH+nVColP4AVCkWebNdQsT1z7922szI3gUgyhrW5KdtUIjxGcy4pKQk+Pj7YtGkTAGDkyJFYunQpoqOj2aYSkvoY1Wc7sr2DpqamqFevHoKDg3XKg4OD0aRJkwzXSUhISPfijIzeTmfMc3RERIVDREQEPv30U2zatAlGRkZYuXIl/Pz8YGJiIndoJCFZT5H4+vqiX79+8PDwQOPGjbWztI0YMQLA29MbT58+xYYNGwAAHTt2xNChQ+Hv7689RTJ+/Hg0aNCAN7shIioE7ty5g9atW+PJkycoXrw4tm/fjlatWskdFuUBWRMMb29vvHz5EnPmzEFERARq1qyJAwcOwNXVFcDbLPfdOTEGDhyIuLg4/Pzzz/j6669RrFgxtGrVCgsXLpTrJRARkR7Kli2LkiVLwtraGvv27UPFihXlDonyiEIY2LmF2NhY2NnZISYmRvJBnpGRkXB0dOS5QwmwPaWRkKJC9RmHAQB/z/ocb16/YptKhMdo9qUNNkwbPPz06VNYW1vDzs4uXT22qXTyoj31+Q6V/SoSIno7higxVfqJrjh5FsktPj4eAwcORN26dTF16lQAQJkyZWSOivIDEwwimQkh0H3VeYQ8ipY7FCJJPX78GJ06dcK1a9ewb98+9O/fH2XLlpU7LMon7IMiklliqjrPkwsP1+KwMDHK030QvevChQuoX78+rl27BgcHB/z5559MLgwMezCICpAr/2sNS1PpEwELEyNeyk35ZuPGjRg6dCiSk5NRq1Yt7N27F25ubnKHRfmMCQZRAWJpagRL07z5t2SCQflh+vTpmDdvHgCgU6dO+O2332BtbS1zVCQHniIhIiLJpE0zMHXqVOzcuZPJhQFjDwYREeXKu/eQ8vHxQd26dVGvXj2ZoyK5sQeDiIhy7OTJk2jUqBGioqK0ZUwuCGCCQUREObRmzRq0bt0aly5dwuzZs+UOhwoYniIhkkBuJsriZFhU2KhUKnz99ddYsWIFgLe3feAtG+h9TDCIcokTZZEhiY6Ohre3t/ZO2HPnzsW0adO0YzCI0jDBIMolqSbK4mRYVNA9ePAA7dq1w927d2FpaYmNGzeia9eucodFBRQTDCIJ5WaiLAsTI/4KpALNzs4OarUaLi4u2Lt3L+rUqSN3SFSAMcEgklBeTpRFJLeSJUviwIEDsLOzg5OTk9zhUAHHq0iIiChDKSkpGD58OFavXq0tq1y5MpMLyhb+1CIionSioqLQvXt3nDx5EmZmZvjiiy/g7Owsd1hUiDDBICIiHbdu3ULHjh0RFhYGGxsbbN68mckF6Y0JBlE2ZDXPBeexoKJk//796N27N+Li4lC+fHns3bsXNWrUkDssKoSYYBB9AOe5IEOxZMkSTJ48GUIItGjRAtu3b4e9vb3cYVEhxUGeRB+Q3XkuOI8FFXYqlQpCCAwfPhxHjhxhckG5wh4MIj1kNc8F57Ggwu6bb75B3bp10aZNGx7LlGvswSDSQ9o8Fxk9+IFMhc3169fRsWNHvHnzBgCgUCjg6enJY5kkwQSDiMgA7dy5E02bNsX+/fsxbdo0ucOhIogJBhGRARFCYO7cuejWrRsSEhLQpk0b3mqd8gTHYBARGYiEhAQMHjwYQUFBAIBx48ZhyZIlMDbmVwFJj0cVEZEBePr0KTp16oSQkBAYGxvDz88PQ4cOlTssKsKYYJBByGqirA/hRFpUFAgh8PTpU5QsWRI7duxAixYt5A6JijgmGFTkcaIsIqBs2bLYv38/SpQoAXd3d7nDIQPAQZ5U5GV3oqwP4URaVJhoNBpMnToV27dv15bVq1ePyQXlG/ZgkEHJaqKsD+FEWlRYxMXFoV+/ftizZw8sLS3RtGlTlC5dWu6wyMAwwSCDkjZRFlFR9fDhQ3h5eeHmzZswNTXFqlWrmFyQLPhJS0RURJw5cwZdu3bFixcv4OTkhN27d6NRo0Zyh0UGimMwiIiKgICAALRq1QovXrxA3bp1cfnyZSYXJCsmGERERcCtW7eQmpqK7t274/Tp03BxcZE7JDJwPEVCRFQELFy4EHXq1EHv3r2hVPK3I8mPRyEVSkIIJKSosvngRFlU9Ny7dw9DhgxBSkoKAMDIyAh9+/ZlckEFBnswqNDhxFlk6I4ePYoePXogOjoa9vb2WLhwodwhEaXDVJcKnZxOnMWJsqgo8PPzg6enJ6Kjo9GwYUOMHz9e7pCIMpSjHgyVSoUTJ07g/v376N27N2xsbPDs2TPY2trC2tpa6hiJMqXPxFmcKIsKs9TUVIwbNw7+/v4AgL59+2LNmjUwNzeXOTKijOmdYDx69Aht27ZFeHg4kpOT8fnnn8PGxgaLFi1CUlISVq1alRdxEmWIE2eRIXj58iV69OiB48ePQ6FQYMGCBZg8eTITZirQ9D5FMm7cOHh4eCA6OhoWFhba8i5duuDo0aOSBkdEREBUVBSuXr0Ka2tr7NmzB9988w2TCyrw9P7pd+bMGZw9exampqY65a6urnj69KlkgRER0VtVqlTBzp074ejoiJo1a8odDlG26N2DodFooFanv+zvyZMnsLGxkSQoIiJDJoTA0qVLdXqFW7VqxeSCChW9E4zPP/8cy5Yt0z5XKBR48+YNZs6cifbt20sZGxGRwUlOTsaQIUPw9ddfo0ePHvjvv//kDokoR/Q+RfLjjz+iZcuWqF69OpKSktC7d2/cu3cP9vb22LJlS17ESKT1doItTpxFRVNkZCS6du2Ks2fPQqlUYtasWXB0dJQ7LKIc0TvBcHZ2xvXr17F161aEhIRAo9FgyJAh6NOnj86gTyKpcYItKsr++usveHl5ITw8HHZ2dti2bRvatGkjd1hEOaZ3gnHq1Ck0adIEgwYNwqBBg7TlKpUKp06dQvPmzSUNkCjN+xNsceIsKip2796Nvn37Ij4+HpUqVcK+fftQpUoVucMiyhW9E4yWLVsiIiIiXbddTEwMWrZsmeEAUCKpXflfa5S0MuWlelQk7NmzB/Hx8WjdujW2bduG4sWLyx0SUa7pnWAIITL8UH/58iWsrKwkCYroQyxNOSsnFR3+/v6oVasWxo4dC2NjThxHRUO2j+SuXbsCeHvVyMCBA2FmZqZdplarcePGDTRp0kT6CImIiphnz55hxYoVmD9/PoyMjGBubg5fX1+5wyKSVLYTDDs7OwBvezBsbGx0BnSampqiUaNGGDp0qPQREhEVIVeuXEGnTp3w7NkzmJubY9asWXKHRJQnsp1gBAQEAADc3NwwceJEng4hItJTUFAQBg4ciKSkJFSvXh39+vWTOySiPKP3RFszZ85kckF5TgiBxFQ1ElJU7zw4gJgKJ41Gg+nTp6NXr15ISkpC+/btcf78eVSoUEHu0IjyTI5GE23fvh3btm1DeHg4UlJSdJZdvXpVksDIcAkh0POXCwgJfy13KES5Fh8fj/79+2Pnzp0AgIkTJ+L777+HkREvsaaiTe8ejBUrVmDQoEFwdHTEtWvX0KBBA5QsWRIPHjxAu3bt8iJGMjCJqeoskwvOf0GFyb1793DgwAGYmpoiMDAQixcvZnJBBkHvHgw/Pz+sXr0aX375JX799VdMnjwZ5cuXx4wZM/Dq1au8iJEM2KWprWBtbqJTZmHCS1Sp8KhTpw5+++03lC5dmlfakUHRuwcjPDxc+09iYWGBuLg4AEC/fv14LxKSnKWpESxNjXUeTC6ooPv1119x5coV7fNu3boxuSCDo3eCUapUKbx8+RIA4OrqigsXLgAAwsLCIISQNjoiokJErVZj4sSJGDhwIDp16oQXL17IHRKRbPROMFq1aoV9+/YBAIYMGYIJEybg888/h7e3N7p06SJ5gEREhUFMTAy8vLzwww8/AHj7+ViyZEmZoyKSj94JxurVqzFt2jQAwIgRIxAYGIhq1aph9uzZ8Pf31zsAPz8/uLu7w9zcHPXq1cPp06ezrJ+cnIxp06bB1dUVZmZmqFChAtavX6/3fomIpHL//n00btwYBw4cgLm5ObZs2YI5c+ZAqdT7I5aoyNB7kKdSqdT5p+nZsyd69uwJAHj69CnKlCmT7W0FBQVh/Pjx8PPzQ9OmTfHLL7+gXbt2CA0NRbly5TJcp2fPnvjvv/+wbt06VKxYEZGRkVCpVPq+DCIiSRw/fhw9e/bEq1ev4OzsjD179sDDw0PusIhkJ8lddZ4/f4758+dj7dq1SExMzPZ6S5cuxZAhQ+Dj4wMAWLZsGQ4fPgx/f38sWLAgXf1Dhw7h5MmTePDgAUqUKAHg7cyiVDilTab1Pk6oRYXJzz//jFevXqF+/frYvXs3nJ2d5Q6JqEDIdoLx+vVrjB49GkeOHIGJiQmmTJmCMWPGYNasWViyZAlq1Kih16mKlJQUhISEYMqUKTrlbdq0wblz5zJcZ+/evfDw8MCiRYuwceNGWFlZwcvLC3PnztW5N8q7kpOTkZycrH0eGxsL4O3MehqNJtvxfohGo4EQQtJtFmXZnUxL6vfJkPEYlVZae65btw5VqlTB9OnTYWFhwfbNBR6j0sqL9tRnW9lOMKZOnYpTp05hwIABOHToECZMmIBDhw4hKSkJBw8eRIsWLfQKMioqCmq1Gk5OTjrlTk5OeP78eYbrPHjwAGfOnIG5uTl27dqFqKgojBo1Cq9evco0uVmwYAFmz56drvzFixdISkrSK+asaDQaxMTEQAjB867Z8KHJtACguqMZ4qJfIoGTEkmCx6g0Xr9+jaCgIPj4+CA2NhZ2dnYYP3484uLitJftU87wGJVWXrSnPsd4thOMP/74AwEBAWjdujVGjRqFihUronLlyli2bFlOYtR6f04DIUSm8xxoNBooFAps2rRJe3fXpUuXonv37li5cmWGvRjffvutzm2QY2Nj4eLiAgcHB9ja2uYq9oxic3Bw4D9GNiSk/P9xM5emtoKlqW4SodFo8Ob1Kzg6OrI9JcJjNPf++ecfdO7cGffu3YOVlRX69OnD9pQQj1Fp5UV7mpubZ7tuthOMZ8+eoXr16gCA8uXLw9zcXDt2Iifs7e1hZGSUrrciMjIyXa9GmtKlS6NMmTLa5AIAqlWrBiEEnjx5gkqVKqVbx8zMDGZmZunK3x+sKgWFQpEn2y2K3m0ja3MTWJrqHooajQYJ/9eWbE/p8BjNucOHD8Pb2xsxMTFwdXVF69at2Z55gG0qLanbU5/tZLumRqOBicn/n7LZyMgoV3dVNTU1Rb169RAcHKxTHhwcnOmMd02bNsWzZ8/w5s0bbdndu3ehVCpRtmzZHMdCRJQZIQSWLVuG9u3bIyYmBp988gkuXbqE2rVryx0aUYGW7R4MIQQGDhyo7Q1ISkrCiBEj0iUZaXcMzA5fX1/069cPHh4eaNy4MVavXo3w8HCMGDECwNvTG0+fPsWGDRsAAL1798bcuXMxaNAgzJ49G1FRUZg0aRIGDx6c6SBPIqKcSklJwejRo7F27VoAwKBBg+Dv7w8zMzMORCT6gGwnGAMGDNB53rdv31zv3NvbGy9fvsScOXMQERGBmjVr4sCBA3B1dQUAREREIDw8XFvf2toawcHB+Oqrr+Dh4YGSJUuiZ8+emDdvXq5jISJ639WrVxEQEAClUoklS5Zg/PjxvBcOUTZlO8EICAjIkwBGjRqFUaNGZbgsMDAwXVnVqlXTnVYhIsoLjRo1gr+/P8qWLYt27drJHQ5RoSLJRFtEQOYTZ2WEk2lRQbVv3z5UrlwZVapUAQAMHTpU5oiICicmGCQJIQS6rzqPkEfRcodClCNCCCxcuBBTp05FpUqVcPHiRRQrVkzusIgKLSYYJInEVHWOkgsP1+KwMOFEWiSvpKQk+Pj4YNOmTQCA1q1b5+oqOSJigkF54Mr/WqebOCszFiZGHDRHsoqIiECXLl1w8eJFGBkZYcWKFZmOCyOi7GOCQZKzNDVKN3EWUUF09epVdOrUCU+ePEHx4sWxfft2tGrVSu6wiIqEHE3ttXHjRjRt2hTOzs549OgRgLd3Qt2zZ4+kwRER5aVvv/0WT548QdWqVXHp0iUmF0QS0jvB8Pf3h6+vL9q3b4/Xr19DrX57NUCxYsVyfV8SIqL8tHHjRgwZMgQXLlxAxYoV5Q6HqEjRO8H46aefsGbNGkybNg1G79zl0sPDAzdv3pQ0OCIiKcXHx2sHcgKAo6Mj1q5dq3N/IyKSht4nysPCwlC3bt105WZmZoiPj5ckKCIiqT1+/BidOnXCtWvXkJqaioEDB8odElGRpncPhru7O65fv56u/ODBg9q7rZLhEEIgIUXFibOoQLtw4QLq16+Pa9euwcHBgadDiPKB3j0YkyZNwujRo5GUlAQhBC5duoQtW7ZgwYIF2hsCkWHg5FpUGGzcuBFDhw5FcnIyatWqhb1798LNzU3usIiKPL0TjEGDBkGlUmHy5MlISEhA7969UaZMGSxfvhy9evXKixipgMpoci1OnEUFhVqtxrRp07Bw4UIAQKdOnfDbb7/B2tpa5siIDEOOJisYOnQohg4diqioKGg0Gjg6OkodFxUyaZNrceIsKijOnj2rTS6mTp2KuXPnQqnM0ZX5RJQDeicYs2fPRt++fVGhQgXY29vnRUxUCHFyLSpomjdvju+++w6urq7o3bu33OEQGRy90/kdO3agcuXKaNSoEX7++We8ePEiL+IiItLb6dOn8fTpU+3zb7/9lskFkUz0TjBu3LiBGzduoFWrVli6dCnKlCmD9u3bY/PmzUhISMiLGImIPmjNmjVo1aoVOnXqxM8iogIgRycka9Soge+++w4PHjzA8ePH4e7ujvHjx6NUqVJSx0dElCWVSoVx48Zh2LBhUKlUqFChgtwhERFymGC8y8rKChYWFjA1NUVqaqoUMRERZUt0dDTat2+PFStWAADmzJmDrVu3wtLSUubIiChHCUZYWBjmz5+P6tWrw8PDA1evXsWsWbPw/PlzqeMjIsrQ3bt30ahRIwQHB8PS0hLbt2/H9OnTeRUTUQGh97D/xo0b49KlS6hVqxYGDRqknQeDiCg/DRs2DHfv3oWLiwv27t2LOnXqyB0SEb1D7wSjZcuWWLt2LWrUqJEX8RARZcuvv/6KsWPHYvXq1XBycpI7HCJ6j94JxnfffZcXcRARZSklJQXHjx+Hp6cnAMDV1RV79uyROSoiyky2EgxfX1/MnTsXVlZW8PX1zbLu0qVLJQmMiChNVFQUunfvjlOnTmHPnj3o2LGj3CER0QdkK8FIu71x2t9ERPnl1q1b6NixI8LCwmBjYwMjI97rhqgwyFaCcfz48Qz/JiLKS/v370fv3r0RFxeH8uXLY+/evRz/RVRI6H2Z6uDBgxEXF5euPD4+HoMHD5YkKCIybEIILF68GF5eXoiLi0OLFi1w8eJFJhdEhYjeCcavv/6KxMTEdOWJiYnYsGGDJEFRwSKEQEKKKoOHWu7QqIg6duwYJk+eDCEEhg0bhiNHjvDmikSFTLavIomNjYUQAkIIxMXFwdzcXLtMrVbjwIEDvG17ESSEQPdV5xHyKFruUMiAfPbZZxg3bhwqVqyI0aNHc/IsokIo2wlGsWLFoFAooFAoULly5XTLFQoFZs+eLWlwJL/EVPUHkwsP1+KwMOHAO8qdv/76C2XLlkXJkiUBAMuWLZM3ICLKlWwnGMePH4cQAq1atcKOHTtQokQJ7TJTU1O4urrC2dk5T4KkguHK/1rD0jR9ImFhYsRfmJQrO3bsQP/+/dGoUSMcOnQIJiYmcodERLmU7QSjRYsWAN7eh6RcuXL8QjFAlqZGsDTVe242okwJITBv3jzMmDEDAGBsbIykpCQmGERFQLa+LW7cuIGaNWtCqVQiJiYGN2/ezLRu7dq1JQuOiIquhIQEDBo0CNu2bQMAjBs3DkuWLIGxMZNYoqIgW//JderUwfPnz+Ho6Ig6depAoVBACJGunkKhgFrNKwuIKGtPnz5Fp06dEBISAhMTE/j5+cHHx0fusIhIQtlKMMLCwuDg4KD9m4gop4QQ8Pb2RkhICOzt7bFjxw40b95c7rCISGLZSjBcXV0z/JuISF8KhQK//PILhg0bht9++w3u7u5yh0REeSBHE2398ccf2ueTJ09GsWLF0KRJEzx69EjS4Eh+GZwJI9KbRqPB5cuXtc9r1KiBM2fOMLkgKsL0TjC+++47WFhYAADOnz+Pn3/+GYsWLYK9vT0mTJggeYAkHyEEeqw6L3cYVMi9efMGXbt2RdOmTXHq1CltOa9EIyra9B6u/fjxY1SsWBEAsHv3bnTv3h3Dhg1D06ZN8emnn0odH8koMVWN0IhYAED10racTIv09ujRI3h5eeHGjRswMzNDRESE3CERUT7RuwfD2toaL1++BAAcOXIErVu3BgCYm5tneI8SKhp+H9GYvzhJL2fOnEH9+vVx48YNODk54cSJE/D29pY7LCLKJ3r3YHz++efw8fFB3bp1cffuXXTo0AEAcOvWLbi5uUkdHxUQzC1IH+vXr8eIESOQmpqKunXrYs+ePXBxcZE7LCLKR3r3YKxcuRKNGzfGixcvsGPHDu19A0JCQvDll19KHiARFS5//vknhgwZgtTUVHTv3h2nT59mckFkgPTuwShWrBh+/vnndOW80RkRAW/vhPrll1+iSpUqmD59OpRKvX/HEFERkKM5eV+/fo1169bh9u3bUCgUqFatGoYMGQI7Ozup4yOiQuD+/fsoVaoUrKysoFAo8NtvvzGxIDJwen8CXLlyBRUqVMCPP/6IV69eISoqCj/++CMqVKiAq1ev5kWMRFSAHT16FPXr18eAAQOg0WgAgMkFEenfgzFhwgR4eXlhzZo12psSqVQq+Pj4YPz48TrXuVPBI4RAYmr27heTkML7ylDWVq5ciXHjxkGtVuPJkyeIi4tjTyYRAchBgnHlyhWd5AJ4e4vlyZMnw8PDQ9LgSFpCCHRfdR4hj6LlDoUKudTUVIwbNw7+/v4AgD59+mDt2rUwNzeXOTIiKij07se0tbVFeHh4uvLHjx/DxsZGkqAobySmqnOUXHi4FuckW6T18uVLeHp6wt/fHwqFAgsWLMDGjRuZXBCRDr17MLy9vTFkyBAsWbIETZo0gUKhwJkzZzBp0iReplqIXPlfa1iaZi9psDAx4iRbBOBtL5iXlxfOnTsHa2trbNq0CV5eXnKHRUQFkN4JxpIlS6BQKNC/f3+oVCoAgImJCUaOHInvv/9e8gApb1iaGsHSNEcXEZEBUygUWLx4MQYPHozff/8dtWrVkjskIiqg9P6GMTU1xfLly7FgwQLcv38fQghUrFgRlpaWeREfEclMCIH79+9r70HUpEkT/P333zrjsIiI3pftMRgJCQkYPXo0ypQpA0dHR/j4+KB06dKoXbs2kwuiIio5ORmDBw9GnTp18Ndff2nLmVwQ0YdkO8GYOXMmAgMD0aFDB/Tq1QvBwcEYOXJkXsZGRDKKjIxEq1atEBgYiMTERM5zQ0R6yfbPkJ07d2LdunXo1asXAKBv375o2rQp1Go1jIx4hUF+0mcui3dxXgvKrr/++gteXl4IDw+HnZ0dgoKC4OnpKXdYRFSIZDvBePz4MZo1a6Z93qBBAxgbG+PZs2e8kVE+4lwWlNd27dqFfv36IT4+HpUqVcK+fftQpUoVucMiokIm2wmGWq2Gqamp7srGxtorSSh/5HQui3dxXgvKTHBwMLp27QoAaN26NbZt24bixYvLHBURFUbZTjCEEBg4cCDMzMy0ZUlJSRgxYgSsrKy0ZTt37pQ2QsqUPnNZvIvzWlBmWrZsiTZt2qBKlSpYunQpB3MSUY5l+9NjwIAB6cr69u0raTCkH85lQVJ4/vw5SpQoAVNTUxgbG2Pfvn3peiuJiPSV7W+ngICAvIyDiGRw5coVdOrUCV988QVWrVoFhULB5IKIJMF7KhMZqKCgIDRr1gzPnj3DmTNnEBsbK3dIRFSEyJ5g+Pn5wd3dHebm5qhXrx5Onz6drfXOnj0LY2Nj1KlTJ28DJCpiNBoNpk+fjl69eiEpKQkdOnTA+fPneZt1IpKUrAlGUFAQxo8fj2nTpuHatWto1qwZ2rVrl+HdWt8VExOD/v3747PPPsunSImKhoSEBPTs2RPz5s0DAEyaNAl79uyBra2tzJERUVEja4KxdOlSDBkyBD4+PqhWrRqWLVsGFxcX+Pv7Z7ne8OHD0bt3bzRu3DifIpWfEAIJKSpOlkU5JoRA3759sWvXLpiamiIwMBCLFi3iRHlElCdkuwQhJSUFISEhmDJlik55mzZtcO7cuUzXCwgIwP379/Hbb79pf4VlJTk5GcnJydrnaeeZNRoNNBpNDqNPT6PRQAgh6TbTCCHQ85cLCAl/nW6febG/giAv29NQCSEwevRoPHz4ENu2bUOTJk3YvrnAY1R6bFNp5UV76rOtHCUYGzduxKpVqxAWFobz58/D1dUVy5Ytg7u7Ozp16pStbURFRUGtVsPJyUmn3MnJCc+fP89wnXv37mHKlCk4ffp0tq/PX7BgAWbPnp2u/MWLF0hKSsrWNrJDo9EgJiYGQggoldJ2DCWmqtMlF7WdrRAX/RJviuh8FnnZnoYmKioK9vb20Gg0+Pjjj3H69GlYWVkhMjJS7tAKNR6j0mObSisv2jMuLi7bdfVOMPz9/TFjxgyMHz8e8+fPh1r9tsu+WLFiWLZsWbYTjDTvT/gkhMhwEii1Wo3evXtj9uzZqFy5cra3/+2338LX11f7PDY2Fi4uLnBwcJD0vLNGo4FCoYCDg4Pk/xgJKf9/ttRLU1vB0tSoyE+WlZftaSjUajW+/fZbBAQE4MKFC3B3d2ebSojHqPTYptLKi/Y0NzfPdl29E4yffvoJa9asQefOnfH9999ryz08PDBx4sRsb8fe3h5GRkbpeisiIyPT9WoAb7OmK1eu4Nq1axgzZgyA/9/9Y2xsjCNHjqBVq1bp1jMzM9OZfTSNUqmU/ABWKBR5st13t2dtbmIwk2vlVXsagtjYWHz55Zc4cOAAgLdTgI8YMYJtKjG2p/TYptKSuj312Y7e31RhYWGoW7duunIzMzPEx8dnezumpqaoV68egoOD0aVLF215cHBwhr0gtra2uHnzpk6Zn58fjh07hu3bt8Pd3V2PV0FUdN2/fx9eXl4IDQ2Fubk5AgMD4e3tzfPaRJSv9E4w3N3dcf36dbi6uuqUHzx4ENWrV9drW76+vujXrx88PDzQuHFjrF69GuHh4RgxYgSAt6c3nj59ig0bNkCpVKJmzZo66zs6OsLc3DxdOZGhOnHiBLp164ZXr17B2dkZe/bsgYeHh9xhEZEB0jvBmDRpEkaPHo2kpCQIIXDp0iVs2bIFCxYswNq1a/Xalre3N16+fIk5c+YgIiICNWvWxIEDB7TJS0RExAfnxCCit44ePYq2bdtCpVKhfv362L17N5ydneUOi4gMlEIIIfRdac2aNZg3bx4eP34MAChTpgxmzZqFIUOGSB6g1GJjY2FnZ4eYmBjJB3lGRkbC0dExTwZ5Vp9xGAAQOsfTIMZg5GV7FlVJSUlo0aIFKlSogHXr1sHCwkJnOdtUWmxP6bFNpZUX7anPd2iOvqmGDh2KoUOHIioqChqNBo6OjjkKlN4SQiAxNfMJtDi5FmUmJiYGNjY2UCqVMDc3R3BwMGxsbIr0FUZEVDjk6qewvb29VHEYLCEEuq86j5BH0XKHQoXMP//8g44dO8Lb21s76Ryn/CaigiJHgzyz+nX04MGDXAVkaBJT1dlOLjxci8PChNM6E3Do0CH06tULMTEx+O233zB58mQmF0RUoOidYIwfP17neWpqKq5du4ZDhw5h0qRJUsVlkK78rzUsTTNPIIr65Fr0YUIILF++HF9//TU0Gg2aNm2KnTt3MrkgogJH7wRj3LhxGZavXLkSV65cyXVAhszS1MggBnBSzqSkpGDUqFFYt24dAGDQoEHw9/fPcCI5IiK5STZMt127dtixY4dUmyOidwgh8MUXX2DdunVQKpX44YcfsG7dOiYXRFRgSZZgbN++HSVKlJBqc0T0DoVCgQEDBsDW1hb79++Hr68vT5cRUYGmd3983bp1dT7YhBB4/vw5Xrx4AT8/P0mDIzJ08fHxsLKyAgD06dMHnp6evHqLiAoFvROMzp076zxXKpVwcHDAp59+iqpVq0oVF5FBE0Jg4cKF8Pf3x8WLF1GqVCkAvDSciAoPvRIMlUoFNzc3eHp6aj/wiEhaSUlJ8PHxwaZNmwAAW7ZswYQJE2SOiohIP3qNwTA2NsbIkSORnJycV/EQGbSIiAh8+umn2LRpE4yMjLBy5UomF0RUKOk9yLNhw4a4du1aXsRCZNBCQkJQv359XLx4EcWLF8eRI0cwatQoucMiIsoRvcdgjBo1Cl9//TWePHmCevXqaQegpaldu7ZkwREZipMnT6Jdu3ZITExEtWrVsHfvXlSsWFHusIiIcizbCcbgwYOxbNkyeHt7AwDGjh2rXaZQKCCEgEKhgFrNG3MR6atOnTpwc3ODm5sbtmzZAjs7O7lDIiLKlWwnGL/++iu+//57hIWF5WU8RAYjOTkZpqamUCgUsLOzw7Fjx+Dg4AAjI95vhogKv2wnGEIIAICrq2ueBUNkKB4/foxOnTqhf//+2vv78MosIipK9BrkyZkDiXLvwoULqF+/Pq5du4aFCxciLi5O7pCIiCSn1yDPypUrfzDJePXqVa4CMjT/1zFEBmLjxo0YOnQokpOTUatWLezduxc2NjZyh0VEJDm9EozZs2dz8JmEhBDoseq83GFQPtBoNJg6dSoWLlwIAOjUqRN+++03WFtbyxwZEVHe0CvB6NWrFxwdHfMqFoOTmKpGaEQsAKB6aVtYmHBwX1EkhED37t2xa9cuAMDUqVMxd+5cKJWS3WuQiKjAyfYnHMdf5K3fRzRmGxdRCoUCLVq0gJmZGTZt2oT58+czuSCiIk/vq0gobzC3KHpUKhWMjd/+i40dOxYdO3ZE+fLlZY6KiCh/ZPtnlEaj4ekRomxau3YtPv74Y7x+/RrA214MJhdEZEjYT0skIZVKhfHjx2Po0KG4efMm1qxZI3dIRESy0PteJESUsejoaHh7eyM4OBgAMGfOHEycOFHmqIiI5MEEg0gCd+/eRceOHXH37l1YWlpiw4YN6Natm9xhERHJhgmGTIQQSEjhjeGKgvPnz6N9+/Z4/fo1XFxcsHfvXtSpU0fusIiIZMUEQwZCCHRfdR4hj6LlDoUkUL58edjY2KBatWrYtWsXnJyc5A6JiEh2TDBkkJiq1kkuPFyLc5KtQkaj0WjnsnBycsLx48dRtmxZmJmZyRwZEVHBwKtIZHblf605yVYhExUVhVatWmHjxo3asgoVKjC5ICJ6BxMMmVmaGjG5KERu3bqFBg0a4OTJk/j666/x5s0buUMiIiqQmGAQZdP+/fvRuHFjhIWFoXz58jh+/DhvVkZElAkmGEQfIITA4sWL4eXlhbi4OLRo0QIXL15EjRo15A6NiKjAYoJBlAUhBAYNGoTJkydDCIFhw4bhyJEjsLe3lzs0IqICjQkGURYUCgVcXFxgZGSEn376CatWrYKpqancYRERFXi8TJUoA0II7eDb2bNno0uXLvj4449ljoqIqPBgDwbRe3bs2IFWrVohMTERAKBUKplcEBHpiQkG0f8RQmDOnDno3r07Tpw4gZ9//lnukIiICi2eIiECkJCQgEGDBmHbtm0AgPHjx2PChAkyR0VEVHgxwSCD9/TpU3Tq1AkhISEwMTGBn58ffHx85A6LiKhQY4JBBu3q1av44osvEBERAXt7e+zYsQPNmzeXOywiokKPCQYZtGLFiiElJQU1a9bE3r174e7uLndIRERFAhMMMmjly5dHcHAwKlasCBsbG7nDISIqMngVCRmUN2/eoFu3bvjjjz+0ZXXr1mVyQUQkMfZgkMF4+PAhvLy8cPPmTZw+fRphYWGwsrKSOywioiKJPRhkEM6cOYP69evj5s2bcHJywt69e5lcEBHlISYYVOStW7cOrVq1QlRUFOrWrYvLly+jUaNGcodFRFSkMcGgIkuj0cDX1xc+Pj5ITU1F9+7dcfr0abi4uMgdGhFRkccEg4ospVKJpKQkAMCsWbMQFBTE0yJERPmEgzypSFu+fDm6deuGzz77TO5QiIgMCnswqEg5duwYvL29oVKpAAAmJiZMLoiIZMAEg4oMPz8/tGnTBtu2bcPy5cvlDoeIyKDxFEk+EUIgMVUNAEhIUcscTdGSmpqKcePGwd/fHwDQt29fjB49WuaoiIgMGxOMfCCEQPdV5xHyKFruUIqcV69eoUePHjh27BgUCgUWLFiAyZMnQ6FQyB0aEZFBY4KRDxJT1RkmFx6uxWFhYiRDREXD7du30bFjR9y/fx/W1tbYtGkTvLy85A6LiIjABCPfXflfa1iavk0qLEyM+Es7F1QqFf777z+4ublh7969qFWrltwhERHR/2GCkc8sTY1gacpml0KtWrWwf/9+VK9eHQ4ODnKHQ0RE7+BVJFRoJCcnY9iwYTh79qy2rEWLFkwuiIgKIP6UpkIhMjISXbp0wblz57B//378+++/sLS0lDssIiLKhOw9GH5+fnB3d4e5uTnq1auH06dPZ1p3586d+Pzzz+Hg4ABbW1s0btwYhw8fzsdoSQ5//fUX6tevj3PnzsHOzg4BAQFMLoiICjhZE4ygoCCMHz8e06ZNw7Vr19CsWTO0a9cO4eHhGdY/deoUPv/8cxw4cAAhISFo2bIlOnbsiGvXruVz5JRfdu/ejaZNmyI8PBwVK1bEhQsX4OnpKXdYRET0AbImGEuXLsWQIUPg4+ODatWqYdmyZXBxcdFOmPS+ZcuWYfLkyahfvz4qVaqE7777DpUqVcK+ffvyOfLsEUIgIUXFibVyQAihvY9IfHw8PvvsM1y8eBFVq1aVOzQiIsoG2cZgpKSkICQkBFOmTNEpb9OmDc6dO5etbWg0GsTFxaFEiRKZ1klOTkZycrL2eWxsrHZdjUaTg8gzj0UIod2mEAI9f7mAkPDX6epJud+iSq1W49atWwCA0aNH44cffoCJiQnbLhfeP0Ypd9ie0mObSisv2lOfbcmWYERFRUGtVsPJyUmn3MnJCc+fP8/WNn744QfEx8ejZ8+emdZZsGABZs+ena78xYsX2lt5S0Gj0SAmJgZCCCiVyreTa72XXNR2tkJc9Eu84dwXH6TRaDBr1iy0a9cOXbp0QXQ0Z0HNrfePUcodtqf02KbSyov2jIuLy3Zd2a8ieX+iKSFEtiaf2rJlC2bNmoU9e/bA0dEx03rffvstfH19tc9jY2Ph4uKiHSgqFY1GA4VCAQcHByiVSiSkqLTLLk1tBUtTI06s9QFXrlzBxo0bsWzZMu1xMHToUH7QSOT9Y5Ryh+0pPbaptPKiPc3NzbNdV7YEw97eHkZGRul6KyIjI9P1arwvKCgIQ4YMwe+//47WrVtnWdfMzAxmZmbpypVKpeQHsEKh0G733W1bm5twcq0P2Lp1KwYNGoSkpCRUrVoVI0eO1GlPkgbbVFpsT+mxTaUldXvqsx3Z3kFTU1PUq1cPwcHBOuXBwcFo0qRJputt2bIFAwcOxObNm9GhQ4e8DpPymEajwfTp0/Hll18iKSkJHTp0QL9+/eQOi4iIcknWn9W+vr7o168fPDw80LhxY6xevRrh4eEYMWIEgLenN54+fYoNGzYAeJtc9O/fH8uXL0ejRo20vR8WFhaws7OT7XVQzrx58wb9+/fHrl27AACTJk3CggULYGRkxEFeRESFnKwJhre3N16+fIk5c+YgIiICNWvWxIEDB+Dq6goAiIiI0JkT45dffoFKpcLo0aMxevRobfmAAQMQGBiY3+FTLjx69AidOnXCX3/9BVNTU6xevRoDBgyQOywiIpKI7AMDRo0ahVGjRmW47P2k4cSJE3kfEOWLJ0+eIDQ0FI6Ojti9ezcaN24sd0hERCQh2ROMokoIuSMo2Jo2bYqgoCDUq1cP5cqVkzscIiKSGIfp5gEhBHqsOi93GAWKWq3GtGnTcPPmTW1Zly5dmFwQERVRTDDyQGKqGqERb2cMrV7aFhYmRjJHJK/Y2Fh4eXnhu+++Q6dOnZCYmCh3SERElMd4iiSP/T6isUFPrnX//n14eXkhNDQU5ubmWLBgASwsLOQOi4iI8hgTjDxmwLkFjh8/ju7du+PVq1dwdnbGnj174OHhIXdYRESUD3iKhPLEL7/8gjZt2uDVq1eoX78+Ll++zOSCiMiAMMEgyWk0Gmzfvh0qlQq9e/fGyZMn4ezsLHdYRESUj3iKhCSnVCqxbds2bN68GaNGjTLoMShERIaKPRgkiX/++Qfz5s3TPi9evDhGjx7N5IKIyECxB4Ny7fDhw/D29kZMTAycnZ0xePBguUMiIiKZsQeDckwIgeXLl6N9+/aIiYlB06ZN8cUXX8gdFhERFQBMMChHUlJSMGzYMIwfPx4ajQaDBg3C0aNH4ejoKHdoRERUAPAUCentxYsX6NatG06fPg2lUonFixdjwoQJHG9BRERaTDBIb1evXsWZM2dga2uLrVu3ol27dnKHREREBQwTDNKbp6cnVq9ejaZNm6JatWpyh0NERAUQx2DQBwkh8OOPPyIsLExb5uPjw+SCiIgyxQSDspSUlIR+/frB19cXXl5eSEpKkjskIiIqBHiKhDIVERGBzp0749KlSzAyMsLIkSNhbm4ud1hERFQIMMGgDIWEhKBTp054+vQpihcvjt9//x2fffaZ3GEREVEhwVMklM62bdvQrFkzPH36FFWrVsWlS5eYXBARkV6YYJAOtVqNH374AYmJiWjbti0uXLiAihUryh0WEREVMkwwSIeRkRF27dqFmTNnYv/+/bCzs5M7JCIiKoSYYBAeP36MNWvWaJ87Oztj1qxZMDIykjEqIiIqzDjI08BduHABnTt3xn///YfixYuje/fucodERERFAHswDNjGjRvx6aef4r///kPt2rVRv359uUMiIqIiggmGAVKr1ZgyZQr69++P5ORkdO7cGWfPnoWrq6vcoRERURHBBMPAxMXFoUuXLli4cCEAYNq0adixYwesra1ljoyIiIoSjsEwMMHBwdi3bx/MzMywfv169O7dW+6Q6APUajVSU1NzvR2NRoPU1FQkJSVBqeRvi9xie0qPbSqtnLanqampJO3PBMPAdO3aFQsWLECrVq3QoEEDucOhLAgh8Pz5c7x+/Vqy7Wk0GsTFxUGhUEiyTUPG9pQe21RaOW1PpVIJd3d3mJqa5mr/TDAMwMaNG9GmTRs4OTkBAKZMmSJzRJQdacmFo6MjLC0tc/2BK4SASqWCsbExP7wlwPaUHttUWjlpT41Gg2fPniEiIgLlypXL1fvABKMIU6lU+Prrr7FixQo0adIEx48fz3VGSvlDrVZrk4uSJUtKsk1+eEuL7Sk9tqm0ctqeDg4OePbsGVQqFUxMTHK8fyYYRVR0dDS8vb0RHBwMAGjXrl2uDhTKX2ljLiwtLWWOhIgMTdoPUbVazQSDdN29excdO3bE3bt3YWlpiY0bN6Jr165yh0U5wF9xRJTfpPrcYYJRxAQHB6Nnz554/fo1XFxcsHfvXtSpU0fusIiIyMDwOqAiRK1WY8KECXj9+jUaN26My5cvM7mgIsnNzQ3Lli3L8fqBgYEoVqyYZPEUJZ9++inGjx+fL/uaPn06hg0bli/7ov+vfv362LlzZ57vhwmGhIQQSEhRISFFLcv+0+6EOnr0aBw/flx71QhRfho4cCA6d+6cp/u4fPlytr+YMkpGvL29cffu3RzvPzAwEEqlUjtfgJOTEzp27Ihbt27leJsFxc6dOzF37tw8389///2H5cuXY+rUqemWnTt3DkZGRmjbtm26ZSdOnIBCocjw8u06depg1qxZOmXXrl1Djx494OTkBHNzc1SuXBlDhw7N1fufHX5+fnB3d4e5uTnq1auH06dPZ1l/4MCBUCgU6R41atTQqbds2TJUqVIFFhYWcHFxwYQJE5CUlKRd7u/vj9q1a8PW1hZ2dnZo1qwZDh48qLON6dOnY8qUKdBoNNK94AwwwZCIEALDtt1BzVnB8Jj3Z77tNyoqSicTrVSpEn7++WeYmZnlWwxE+c3BwSFXA2AtLCzg6OiYqxhsbW0RHh6Op0+f4o8//kB8fDw6dOiAlJSUXG33Q6SYdC0rJUqUgI2NTZ7uAwDWrVuHxo0bw83NLd2y9evX46uvvsKZM2cQHh6e433s378fjRo1QnJyMjZt2oTbt29j48aNsLOzw/Tp03MRfdaCgoIwfvx4TJs2DdeuXUOzZs3Qrl27LF/L8uXLERERoX08fvwYJUqUQI8ePbR1Nm3ahClTpmDmzJm4ffs21q1bh6CgIHz77bfaOmXLlsX333+PK1eu4PLly/j000/RuXNnneS3Q4cOiImJweHDh/OmAdIIAxMTEyMAiJiYGEm3G5eYLFy/2a/z6OZ3Vmg0Gkn3866///5buLu7C6VSKQ4fPpxn+5GDWq0WERERQq1Wyx2KLBITE0VoaKhITEzUlmk0GhGfnJrjx5ukFPH6TYJ4k5Si97r6HMcDBgwQnTp1ynT5iRMnRP369YWpqakoVaqU+Oabb0Rqaqp2eWxsrOjdu7ewtLQUpUqVEkuXLhUtWrQQ48aN09ZxdXUVP/74o/b5zJkzhYuLizA1NRWlS5cWX331lRBCiBYtWggAOg8hhAgICBB2dnY6ce3Zs0fUq1dPmJmZiZIlS4ouXbpk+hrS1k9JSdG2zd69ewUAcePGDW29s2fPimbNmglzc3NRtmxZ8dVXX4k3b95olz979ky0b99emJubCzc3N7Fp06Z0rw2A8Pf3F15eXsLS0lLMmDFDu7+PP/5YmJmZCXd3dzFr1iyddsysTYQQYuXKlaJixYrCzMxMODo6im7dummXvd/Wr169Ev369RPFihUTFhYWom3btuLu3bvp2uLQoUOiatWqwsrKSnh6eopnz55l2n5CCFGrVi3x888/65RpNBoRHR0tbGxsxD///CO8vb3F7NmzdeocP35cABDR0dHptvnRRx+JmTNnCiGEiI+PF/b29qJz584Z7j+j9aXSoEEDMWLECJ2yqlWriilTpmR7G7t27RIKhUI8fPhQWzZ69GjRqlUrnXq+vr7ik08+yXAbGo1GpKSkiOLFi4u1a9fqLBs4cKDo169fhutl9PmTRp/vUA7yzANX/tcalqZGsDAxyrOrAPbv34/evXsjLi4O5cuXR5kyZfJkP1RwJKaqUX1GHv/iyEToHE9Ymub+4+Lp06do3749Bg4ciA0bNuCff/7B0KFDYW5uru3a9vX1xdmzZ7F37144OTlhxowZuHr1aqbjibZv344ff/wRW7duRY0aNfD8+XP89ddfAN5293/00UcYNmwYhg4dmmlcf/zxB7p27Ypp06Zh48aNSElJwR9//JHt1/X69Wts3rwZALSX9d28eROenp6YO3cu1q1bhxcvXmDMmDEYM2YMAgICAAD9+/dHVFQUTpw4ARMTE/j6+iIyMjLd9mfOnIkFCxbgxx9/hJGREQ4fPoy+fftixYoVaNasGe7fv689ZTRz5sws2+TKlSsYO3YsNm7ciCZNmuDVq1dZdt8PHDgQ9+7dw969e2Fra4tvvvkG7du3R2hoqPa1JiQkYMmSJdi4cSOUSiX69u2LiRMnYtOmTRluMzo6Gn///Tc8PDzSLfv9999RpUoVVKlSBX379sVXX32F6dOn6/1ZevjwYURFRWHy5MkZLs9qDM6IESPw22+/Zbn90NBQlCtXLl15SkoKQkJC0k1o2KZNG5w7d+7Dgf+fdevWoXXr1jo3ofzkk0/w22+/4dKlS2jQoAEePHiAAwcOYMCAARluQ61WIygoCPHx8WjcuLHOsgYNGmDRokXZjicnmGDkAUtTI0k+jDMihMCSJUvwzTffQAiBFi1aYPv27bC3t8+T/RFJyc/PDy4uLvj555+hUChQtWpVPHv2DN988w1mzJiB+Ph4/Prrr9i8eTM+++wzAEBAQACcnZ0z3WZ4eDhKlSqF1q1bw8TEBOXKldNOg1+iRAkYGRnBxsYGpUqVynQb8+fPR69evTB79mxt2UcffZTla4mJiUHx4sXfjr1KSAAAeHl5oWrVqgCAxYsXo3fv3toBk5UqVcKKFSvQokUL+Pv74+HDh/jzzz9x+fJl7Rft2rVrUalSpXT76t27NwYPHqx93q9fP0yZMkX7xVK+fHnMnTsXkydPxsyZM7Nsk/DwcFhZWeGLL76AjY0NXF1dUbdu3QxfY1picfbsWTRp0gTA2256FxcX7N69W9t9n5qailWrVqFChQoAgDFjxmDOnDmZtt2jR48ghMjwfQ0ICECfPn0AAG3btsWbN29w9OhRtG7dOtPtZRY7AO37oY85c+Zg4sSJWdbJ7JiMioqCWq1ONwbOyckJz58/z9b+IyIicPDgQW3SmqZXr1548eIFPvnkE+0kWiNHjkyXzNy8eRONGzdGUlISrK2tsXPnTlSvXl2nTpkyZRAeHg6NRpNn931hglGIJCUlYfjw4diwYQMAYNiwYfjpp584O6eBsDAxQugczxyvL3IxS6KFiVGO9/uu27dvo3Hjxjr7b9q0Kd68eYMnT54gOjoaqampOvfJsbOzQ5UqVTLdZo8ePbBs2TKUL18ebdu2Rfv27dGxY0cYG2f/4+369etZ9nBkxMbGBhcvXgQAnDp1CosXL8aqVau0y0NCQvDvv//q/IoX/3dviLCwMNy9exfGxsb4+OOPtcsrVqyI4sWLp9vX+7/0Q0JCcPnyZcyfP19bplarkZSUhISEhCzb5PPPP4erq6t2Wdu2bdGlS5cMx7Tcvn0bxsbGaNiwobasZMmSqFKlCm7fvq0ts7S01CYXAFC6dOkMe2LSJCYmAgDMzc11yu/cuYPLly9rx5UZGxvD29sb69ev1zvBEELoVf9djo6OuR6j8/7/mBAi2/93aVc5vT9Y+sSJE5g/fz78/PzQsGFD/Pvvvxg3bhxKly6tM6akSpUquH79OqKjo/H7779j4MCBOHnypE6SYWFhAY1Gg+TkZFhYWOT8hWaBCUYhsmPHDmzYsAFGRkZYtmwZRo8ezYmYDIhCochVz5gQAiolZJ2GOaMP2bQvAoVCofN3RnUy4uLigjt37iA4OBh//vknRo0ahcWLF+PkyZPZnoUwJx+wSqUSFStWhLGxMapVq4bnz5/D29sbp06dAvD2ng7Dhw/H2LFj061brlw53LlzJ8PtZvRaraysdJ5rNBrMnj07wwn0zM3Ns2wTGxsbXL16FSdOnMCRI0cwY8YMzJo1C5cvX0532iCzdn//fXy/nd99LzOS1uMaHR0NBwcHbfm6deugUqlQtmxZnX2ZmJggOjoaxYsXh62tLYC3PUjvx/v69WvY2dkBACpXrgwA+Oeff9KdHviQ3Jwisbe3h5GRUbreisjIyGxd2SeEwPr169GvX790Px6nT5+Ofv36wcfHBwBQq1YtxMfHY9iwYZg2bZq2J8LU1BQVK1aEEAJ16tTB1atXsXz5cvzyyy/abb169QqWlpZ5llwAvIqkUOnduzcmTJiAgwcPYsyYMUwuqNCpXr06zp07p/Plc+7cOdjY2KBMmTKoUKECTExMcOnSJe3y2NhYbXd3ZiwsLODl5YUVK1bgxIkTOH/+PG7evAng7YetWp31peO1a9fG0aNHc/HKgAkTJuCvv/7Crl27AAAff/wxbt26hYoVK6Z7mJqaomrVqlCpVLh27Zp2G//++2+27p778ccf486dOxluO+1LJqs2MTY2RuvWrbFo0SLcuHEDDx8+xLFjx9Ltp3r16lCpVNqeGgB4+fIl7t69i2rVquW4rSpUqABbW1uEhoZqy1QqFTZu3IhFixbh2rVruH79Oq5fv46//voLrq6u2p6gSpUqQalU4vLlyzrbjIiIwNOnT7W9XW3atIG9vX2m4wyyauc5c+Zo95/ZI7NTJKampqhXr572Ng1pgoODtaeZsnLy5En8+++/GDJkSLplCQkJ6U5nGBkZQQiRZUInhEBycrJO2d9//63Te5YX2INRwB04cACffPIJbG1toVAosHTpUrlDIvqgmJgYXL9+XaesRIkSGDVqFJYtW4avvvoKY8aMwZ07dzBz5kz4+vpCqVTCxsYGAwYMwKRJk1CiRAk4Ojpi5syZUCqVmSbUgYGBUKvVaNiwoXZqfAsLC+3gODc3N5w6dQq9evWCmZlZhuOVZs6cic8++wwVKlRAr169oFKpcPDgwUwHCGbE1tYWPj4+mDlzJjp37oxvvvkGjRo1wujRozF06FBYWVnh9u3bCA4Oxk8//YSqVauidevWGDZsGPz9/WFiYoKvv/4aFhYWH/zxMGPGDHzxxRdwcXFBjx49oFQqcePGDdy8eRPz5s3Lsk3279+PBw8eoHnz5ihevDgOHDgAjUaT4WmoSpUqoVOnThg6dCh++eUX2NjYYMqUKShTpgw6deqU7bZ5n1KpROvWrXHmzBntaYD9+/cjOjoagwYNQsmSJXXaoHv37li3bh3GjBkDGxsbDB8+HF9//TWMjY3x0Ucf4dmzZ5g2bRqqVauGNm3aAHjb67N27Vr06NEDXl5eGDt2LCpWrIioqChs27YN4eHh2Lp1a4bx5fYUia+vL/r16wcPDw80btwYq1evRnh4OEaMGKGt8+233+Lp06faU95p1q1bh4YNG6JmzZrpttuxY0csXboUdevW1Z4imT59Ory8vGBk9PY05tSpU9GuXTu4uLggNjYWmzdvxokTJ3Do0CGdbZ0+fVrbVnnmg9eZFDH5cZlqfHLqh1f4AI1GI+bMmSMAiPbt2wuVSiVBlIUHL1PN/DKxnEq7ZC0vL50W4u1lqnjv0lAAYsCAAUKInF2m2qBBA51L/N69lHPXrl2iYcOGwtbWVlhZWYlGjRqJP//8U1v3/Pnzonbt2sLMzCzLy1R37Ngh6tSpI0xNTYW9vb3o2rVrpq8xo8tUhRDi0aNHwtjYWAQFBQkhhLh06ZL4/PPPhbW1tbCyshK1a9cW8+fP19Z/9uyZaNeunTAzMxOurq5i8+bNwtHRUaxatUpbB4DYtWtXuhgOHTokmjRpIiwsLIStra1o0KCBWL169Qfb5PTp06JFixaiePHiwsLCQtSuXVsbrxCZX6ZqZ2cnLCwshKenZ4aXqb5r165d4kNfL4cOHRJlypTR/o9/8cUXon379hkeoyEhIQKACAkJEUIIkZSUJObMmSOqVasmLCwshKurqxg4cKCIiIhIt5/Lly+Lrl27CgcHB2FmZiYqVqwohg0bJu7du5dlfLm1cuVK4erqKkxNTcXHH38sTp48qbN8wIABokWLFjplr1+/FhYWFtr38X2pqali1qxZokKFCsLc3Fy4uLiIUaNG6VxyO3jwYO1+HRwcRKtWrdJNY/DkyRNhYmIiHj9+nOF+pLpMlQmGRKRMMOLj40XPnj21H8zjxo3T+QA2BEwwCm+CIbU3b94IOzu7dNfxyy0v2vPx48cCgE6CVFRpNBrRoEEDsXnzZp2ywniMFlSZtefEiRPF0KFDM12P82AUUU+fPkWnTp0QEhICY2Nj+Pn56T26nagwu3btGv755x80aNAAMTEx2ssdc9MlX1AdO3YMb968Qa1atRAREYHJkyfDzc0NzZs3lzu0PKdQKLB69WrcuHFD7lAMjqOj4wcvw5UCE4wC5OLFi+jcuTOeP3+OkiVLYseOHWjRooXcYRHluyVLluDOnTvaAXOnT58uknO9pKamYurUqXjw4AFsbGzQpEkTbNq0KdtXvxR2H3300QfnGyHpTZo0KV/2wwSjgFCpVOjbty+eP3+OmjVrYu/evXB3d5c7LKJ8V7duXYSEhMgdRr7w9PSEp2fO5zYhKsh4mWoBYWxsjKCgIHh7e+PcuXNMLoiIqFBjgiGjuLg4nDhxQvv8448/xtatW/PlToZERER5iQmGTB4+fIimTZuibdu2OpMKERERFQVMMCSiz7T3Z86cQf369XHz5k0UK1YsV3PmExERFURMMCQghID36gvZqrt+/Xq0atUKUVFRqFu3Li5fvqxzIyEiIqKigAmGBBJT1QiNiAMAVC9tk+GdJ9VqNb7++msMGTIEqamp6N69O06fPg0XF5f8DpeIiCjPMcGQWNCwRhneR2DDhg3a+4jMmjULQUFB6e6QSES55+bmhmXLlskdRqbyK76HDx9CoVDo3BPm7NmzqFWrFkxMTNC5c2ecOHECCoUiWzdYI9IXEwyJZXaPogEDBuDLL7/Etm3btDdvIiqKBg4cCIVCAYVCAWNjY5QrVw4jR45EdHS03KHludjYWEybNg1Vq1aFubk5SpUqhdatW2Pnzp35PtbKxcUFEREROjfN8vX1RZ06dRAWFobAwEA0adIEERER2lucE0mJE23lofPnz6Nu3bowNzeHUqnE5s2b5Q6JKF+0bdsWAQEBUKlUCA0NxeDBg/H69Wts2bJF7tDyzOvXr/HJJ58gJiYG8+bNQ/369WFsbIyTJ09i8uTJaNWqFYoVK5Zv8RgZGaFUqVI6Zffv38eIESNQtmxZbdn7dfSVkpICU1PTXG2Diib+jM4jfn5+aNasGYYOHcqrREhS8fHxmT6SkpKyXTcxMTFbdXPCzMwMpUqVQtmyZdGmTRt4e3vjyJEj2uVqtRpDhgyBu7s7LCwsUKVKFSxfvlxnGwMHDkTnzp2xZMkSlC5dGiVLlsTo0aORmpqqrRMZGYmOHTvCwsIC7u7u2LRpU7pYwsPD0alTJ1hbW8PW1hY9e/bEf//9p10+a9Ys1KlTB+vXr0e5cuVgbW2NkSNHQq1WY9GiRShVqhQcHR0xf/78LF/z1KlT8fDhQ1y8eBEDBgxA9erVUblyZQwdOhTXr1+HtbV1hustXboUtWrVgpWVFVxcXDBq1Ci8efNGu/zRo0fo2LEjihcvDisrK9SoUQMHDhwAAERHR6NPnz5wcHCAhYUFKlWqhICAAAC6p0jS/n758iUGDx4MhUKBwMDADE+RnDt3Ds2bN4eFhQVcXFwwduxYnePAzc0N8+bNw8CBA2FnZ8d7JVGmZE8w/Pz84O7uDnNzc+09B7Jy8uRJ1KtXD+bm5ihfvjxWrVqVT5FmT2pqKkaNGoXRo0dDrVYDeDsNOJFUrK2tM31069ZNp66jo6N2mY2NDYoXLw4bGxtYW1ujXbt2OnXd3Nwy3GZuPXjwAIcOHdK5v4ZGo0HZsmWxbds2hIaGYsaMGZg6dSq2bdums+7x48dx//59HD9+HL/++isCAwMRGBioXT5w4EA8fPgQx44dw/bt2+Hn54fIyEjtciEEOnfujFevXuHkyZMIDg7G/fv34e3trbOf+/fv4+DBgzh06BC2bNmC9evXo0OHDnjy5AlOnjyJhQsX4n//+x8uXMj4ajGNRoOtW7eiT58+cHZ2Trfc2toaxsYZdxgrlUqsWLECf//9N3799VccO3YMkydP1i4fPXo0kpOTcerUKdy8eRMLFy7Uvi/Tp09HaGgoDh48iNu3b8Pf3z/De7aknS6xtbXFsmXLEBERka4NAODmzZvw9PRE165dcePGDQQFBeHMmTMYM2aMTr3FixejZs2aCAkJwfTp0zN8XUSy3q5969atwsTERKxZs0aEhoaKcePGCSsrK/Ho0aMM6z948EBYWlqKcePGidDQULFmzRphYmIitm/fnu195sXt2uOTU4XrN/tF2bGbRfMWnwoAQqFQiO+//563Hc4h3q4989slA8j00b59e526lpaWmdZt0aKFTl17e/sM6+lrwIABwsjISFhZWQlzc3PtdpYuXZrleqNGjRLdunXT2Y6rq6tQqVTash49eghvb28hhBB37twRAMSFCxe0y2/fvi0AiB9//FEIIcSRI0eEkZGRCA8P19a5deuWACAuXbokhBBi5syZwtLSUsTGxmrreHp6Cjc3N53jr0qVKmLBggXa5+/eCvu///7L1msUQghXV1dtfBnZtm2bKFmypPZ5rVq1xKxZszKs27FjRzFo0KAMl4WFhQkA4tq1a9oyOzs7ERAQoH1+/PhxAUBER0cLIYTo16+fGDZsmM52Tp8+LZRKpfZYdHV1FZ07d87iFeYcb9curZy2Z5G4XfvSpUsxZMgQ+Pj4AACWLVuGw4cPw9/fHwsWLEhXf9WqVShXrpx2BHa1atVw5coVLFmyJN0vt/yWGvUYkTvm4MnrCFhbW2Pz5s3o2LGjrDFR0fRu9/n7jIx0L5F+/9e8SqWCsbExFApFuoHGDx8+lCzGli1bwt/fHwkJCVi7di3u3r2Lr776SqfOqlWrsHbtWjx69AiJiYlISUlBnTp1dOrUqFFD5zWVLl0aN2/eBADcvn0bxsbG8PDw0C6vWrWqzjiH27dvw8XFRedy8OrVq6NYsWK4ffs26tevD+Bt7827U/Q7OTnByMhIp42cnJx02vNd4v9Og2Z0BdmHHD9+HN999x1CQ0MRGxsLlUqFpKQkxMfHw8rKCmPHjsXIkSNx5MgRtG7dGt26dUPt2rUBACNHjkS3bt1w9epVtGnTBp07d0aTJk30jiFNSEgI/v33X51TTUIIaDQahIWFoVq1agCg0+ZEmZEtwUhJSUFISAimTJmiU96mTRucO3cuw3XOnz+PNm3a6JR5enpi3bp1SE1NzfAWx8nJyUhOTtY+j42NBfC2S1Oj0eT2ZWj3Ebl9FlQx/6Gcqyv27d2LmjVrSrZ9Q6TRaLQfbIYo7fWnPd5laWmZ5brv1n+/7vv/J1nVzahOdllZWaFChQoAgOXLl6NVq1aYNWsW5s6dCwDYtm0bJkyYgCVLlqBx48awsbHB4sWLcenSJZ39mZiYpNt/RsfG+3XS2k2j0UChUGS4/N167+9HoVBkWKZWq3XK0v62t7dH8eLFERoamq32Stvvo0eP0L59ewwfPhxz5sxBiRIlcObMGfj4+CAlJQWWlpYYMmQI2rRpgz/++APBwcFYsGABlixZgq+++gpt27bFw4cP8ccff+Do0aP47LPPMGrUKCxZsiTda3x/3xm1g0ajwbBhwzB27Nh0MZcrV05b39LSMs/Glr0fG+VOTtrz3ePh/c9gfT6TZUswoqKioFar4eTkpFPu5OSE58+fZ7jO8+fPM6yvUqkQFRWF0qVLp1tnwYIFmD17drryFy9epBsQl1OJqWqUbDcOMeeD8PvOQDg6Omb6S4eyR6PRICYmBkIIg7ykNzU1FRqNBiqVSrIxPEII7bignPzSzq60D6V34542bRo6duyIoUOHwtnZGadOnULjxo0xbNgwbZ379+9re1ky207aB59KpUKlSpWgUqlw8eJFbU/EnTt38Pr1a+16VapUQXh4OMLCwrS9GKGhoYiJiUHlypWhUqm0Ccu7+/nQvtOev9uePXr0wKZNmzBt2rR04zDi4+NhZmamHYeRtu2LFy9CpVJh4cKF2uN869atAKDz3pcuXRo+Pj7w8fHBtGnTsGbNGowcORIAULx4cfTt2xd9+/ZFkyZNMGXKFHz//ffadd8/ht59Xe+OE1OpVKhTpw5u3boFNze3DN/bd9+bvBhbll/HqKHIaXum/V+8fPky3Q/3uLi4bG9H9stU33/RQogsGyKj+hmVp/n222/h6+urfR4bGwsXFxc4ODjA1tY2p2Gni+Huugl48aIvyjmXStdNTfpL++Xp4OBgkAlGUlIS4uLiYGxsnOngwJzKqKdPSkqlEkqlUifuzz77DDVq1MCiRYvw888/o1KlSvjtt99w9OhRuLu7Y+PGjbhy5Qrc3d2162W0nXfn16hRowbatm2LkSNH4pdffoGxsTEmTJgACwsL7Xqenp6oXbs2Bg4ciB9//BEqlQqjR49GixYttFP0K5VK7Tazeg3v7vtdae25YMECnDp1Cp988gnmzZsHDw8PmJiY4PTp0/j+++9x6dIl7embtG2nJTn+/v7o2LEjzp49izVr1gCA9r0fP3482rVrh8qVKyM6OhonT55E9erVYWxsjBkzZqBevXqoUaMGkpOTcfDgQVSrVk3nuHn/GHr3daV9VqXVmTJlCho3boxx48Zh6NChsLKywu3btxEcHIyffvopw23khbw+Rg2Nvu1pbGwMpVKJkiVLwtzcXGfZ+8+z3I5ee5WQvb09jIyM0vVWREZGpuulSFOqVKkM6xsbG6NkyZIZrmNmZgYzM7N05WkfIFKxNlcgwcwk3Xlbyrm0cQKG2J5pX3ppDym8m7znx6/D9/fh6+uLQYMGYcqUKRg5ciT++usv9OrVCwqFAl9++SVGjRqFgwcPplvv3efvxx8QEAAfHx98+umncHJywrx58zB9+nSdttu9eze++uortGjRAkqlEm3btsVPP/2UblsZtUlGsaSVvd+eJUqUwIULF/D9999j/vz5ePToEYoXL45atWph8eLFKFasmE59hUKBunXrYunSpVi0aBGmTp2K5s2bY8GCBejfv7+2jkajwZgxY/DkyRPY2tqibdu2+PHHH6FQKGBmZqa9PNbCwgLNmjXD1q1bdeJ8/xh6f9m7ZR999BFOnjyJadOmoXnz5hBCoEKFCvD29s50G1LK72O0qMtpe6a9vxl9/urzeawQMp7oatiwIerVqwc/Pz9tWfXq1dGpU6cMB3l+88032LdvH0JDQ7VlI0eOxPXr13H+/Pls7TM2NhZ2dnaIiYmRrAcDePuLOzIyEo6Ojgb5hSg1Q2/PpKQkhIWFaS/hlsL7gzwpd9ie0mObSiun7ZnV548+36GyfnL7+vpi7dq1WL9+PW7fvo0JEyYgPDwcI0aMAPD29Eb//v219UeMGIFHjx7B19cXt2/fxvr167Fu3TpMnDhRrpdAREREGZB1DIa3tzdevnyJOXPmaOfMP3DgAFxdXQEAERERCA8P19Z3d3fHgQMHMGHCBKxcuRLOzs5YsWKF7JeoEhERkS5ZT5HIgadICgdDb0+eIin42J7SY5tKy6BPkRAREVHRxASDqAAzsA5GIioApPrcYYJBVAClXbeekJAgcyREZGhSUlIApL/1gL5kn2iLiNIzMjJCsWLFtDPCWlpa5vqcNM9vS4vtKT22qbRy0p4ajQYvXryApaVlridTY4JBVECVKlUKACSbdj7t3gJpk3hR7rA9pcc2lVZO21OpVKJcuXK5fg+YYBAVUAqFAqVLl4ajoyNSU1Nzvb20ewuULFnSIK/MkRrbU3psU2nltD1NTU0laX8mGEQFnJGRkST3t9FoNDAxMYG5uTk/vCXA9pQe21Racrcn30EiIiKSHBMMIiIikhwTDCIiIpKcwY3BSJtAJDY2VtLtajQaxMXF8dyhRNie0mObSovtKT22qbTyoj3TvjuzMxmXwSUYcXFxAAAXFxeZIyEiIiqc4uLiYGdnl2Udg7vZmUajwbNnz2BjYyPpddaxsbFwcXHB48ePJb2JmqFie0qPbSottqf02KbSyov2FEIgLi4Ozs7OH+wVMbgeDKVSibJly+bZ9m1tbfmPISG2p/TYptJie0qPbSotqdvzQz0XaXiSi4iIiCTHBIOIiIgkxwRDImZmZpg5cybMzMzkDqVIYHtKj20qLban9Nim0pK7PQ1ukCcRERHlPfZgEBERkeSYYBAREZHkmGAQERGR5JhgEBERkeSYYGSTn58f3N3dYW5ujnr16uH06dNZ1j958iTq1asHc3NzlC9fHqtWrcqnSAsPfdp0586d+Pzzz+Hg4ABbW1s0btwYhw8fzsdoCz59j9E0Z8+ehbGxMerUqZO3ARZC+rZpcnIypk2bBldXV5iZmaFChQpYv359PkVbOOjbpps2bcJHH30ES0tLlC5dGoMGDcLLly/zKdqC7dSpU+jYsSOcnZ2hUCiwe/fuD66Tr99Ngj5o69atwsTERKxZs0aEhoaKcePGCSsrK/Ho0aMM6z948EBYWlqKcePGidDQULFmzRphYmIitm/fns+RF1z6tum4cePEwoULxaVLl8Tdu3fFt99+K0xMTMTVq1fzOfKCSd/2TPP69WtRvnx50aZNG/HRRx/lT7CFRE7a1MvLSzRs2FAEBweLsLAwcfHiRXH27Nl8jLpg07dNT58+LZRKpVi+fLl48OCBOH36tKhRo4bo3LlzPkdeMB04cEBMmzZN7NixQwAQu3btyrJ+fn83McHIhgYNGogRI0bolFWtWlVMmTIlw/qTJ08WVatW1SkbPny4aNSoUZ7FWNjo26YZqV69upg9e7bUoRVKOW1Pb29v8b///U/MnDmTCcZ79G3TgwcPCjs7O/Hy5cv8CK9Q0rdNFy9eLMqXL69TtmLFClG2bNk8i7Gwyk6Ckd/fTTxF8gEpKSkICQlBmzZtdMrbtGmDc+fOZbjO+fPn09X39PTElStXkJqammexFhY5adP3pd2GuESJEnkRYqGS0/YMCAjA/fv3MXPmzLwOsdDJSZvu3bsXHh4eWLRoEcqUKYPKlStj4sSJSExMzI+QC7yctGmTJk3w5MkTHDhwAEII/Pfff9i+fTs6dOiQHyEXOfn93WRwNzvTV1RUFNRqNZycnHTKnZyc8Pz58wzXef78eYb1VSoVoqKiULp06TyLtzDISZu+74cffkB8fDx69uyZFyEWKjlpz3v37mHKlCk4ffo0jI35MfC+nLTpgwcPcObMGZibm2PXrl2IiorCqFGj8OrVK47DQM7atEmTJti0aRO8vb2RlJQElUoFLy8v/PTTT/kRcpGT399N7MHIpvdv7S6EyPJ27xnVz6jckOnbpmm2bNmCWbNmISgoCI6OjnkVXqGT3fZUq9Xo3bs3Zs+ejcqVK+dXeIWSPseoRqOBQqHApk2b0KBBA7Rv3x5Lly5FYGAgezHeoU+bhoaGYuzYsZgxYwZCQkJw6NAhhIWFYcSIEfkRapGUn99N/OnyAfb29jAyMkqXYUdGRqbLBNOUKlUqw/rGxsYoWbJknsVaWOSkTdMEBQVhyJAh+P3339G6deu8DLPQ0Lc94+LicOXKFVy7dg1jxowB8PbLUQgBY2NjHDlyBK1atcqX2AuqnByjpUuXRpkyZXRuZV2tWjUIIfDkyRNUqlQpT2Mu6HLSpgsWLEDTpk0xadIkAEDt2rVhZWWFZs2aYd68eQbfG6yv/P5uYg/GB5iamqJevXoIDg7WKQ8ODkaTJk0yXKdx48bp6h85cgQeHh4wMTHJs1gLi5y0KfC252LgwIHYvHkzz8G+Q9/2tLW1xc2bN3H9+nXtY8SIEahSpQquX7+Ohg0b5lfoBVZOjtGmTZvi2bNnePPmjbbs7t27UCqVKFu2bJ7GWxjkpE0TEhKgVOp+TRkZGQH4/7+8Kfvy/bspT4aOFjFpl1atW7dOhIaGivHjxwsrKyvx8OFDIYQQU6ZMEf369dPWT7sUaMKECSI0NFSsW7eOl6m+R9823bx5szA2NhYrV64UERER2sfr16/legkFir7t+T5eRZKevm0aFxcnypYtK7p37y5u3bolTp48KSpVqiR8fHzkegkFjr5tGhAQIIyNjYWfn5+4f/++OHPmjPDw8BANGjSQ6yUUKHFxceLatWvi2rVrAoBYunSpuHbtmvayX7m/m5hgZNPKlSuFq6urMDU1FR9//LE4efKkdtmAAQNEixYtdOqfOHFC1K1bV5iamgo3Nzfh7++fzxEXfPq0aYsWLQSAdI8BAwbkf+AFlL7H6LuYYGRM3za9ffu2aN26tbCwsBBly5YVvr6+IiEhIZ+jLtj0bdMVK1aI6tWrCwsLC1G6dGnRp08f8eTJk3yOumA6fvx4lp+Lcn838XbtREREJDmOwSAiIiLJMcEgIiIiyTHBICIiIskxwSAiIiLJMcEgIiIiyTHBICIiIskxwSAiIiLJMcEgIiIiyTHBICpiAgMDUaxYMbnDyDE3NzcsW7YsyzqzZs1CnTp18iUeIsoZJhhEBdDAgQOhUCjSPf7991+5Q0NgYKBOTKVLl0bPnj0RFhYmyfYvX76MYcOGaZ8rFArs3r1bp87EiRNx9OhRSfaXmfdfp5OTEzp27Ihbt27pvZ3CnPAR5RQTDKICqm3btoiIiNB5uLu7yx0WgLd3ZI2IiMCzZ8+wefNmXL9+HV5eXlCr1bnetoODAywtLbOsY21tnSe3l37fu6/zjz/+QHx8PDp06ICUlJQ83zdRYccEg6iAMjMzQ6lSpXQeRkZGWLp0KWrVqgUrKyu4uLhg1KhROrcIf99ff/2Fli1bwsbGBra2tqhXrx6uXLmiXX7u3Dk0b94cFhYWcHFxwdixYxEfH59lbAqFAqVKlULp0qXRsmVLzJw5E3///be2h8Xf3x8VKlSAqakpqlSpgo0bN+qsP2vWLJQrVw5mZmZwdnbG2LFjtcvePUXi5uYGAOjSpQsUCoX2+bunSA4fPgxzc3O8fv1aZx9jx45FixYtJHudHh4emDBhAh49eoQ7d+5o62T1fpw4cQKDBg1CTEyMtidk1qxZAICUlBRMnjwZZcqUgZWVFRo2bIgTJ05kGQ9RYcIEg6iQUSqVWLFiBf7++2/8+uuvOHbsGCZPnpxp/T59+qBs2bK4fPkyQkJCMGXKFJiYmAAAbt68CU9PT3Tt2hU3btxAUFAQzpw5gzFjxugVk4WFBQAgNTUVu3btwrhx4/D111/j77//xvDhwzFo0CAcP34cALB9+3b8+OOP+OWXX3Dv3j3s3r0btWrVynC7ly9fBgAEBAQgIiJC+/xdrVu3RrFixbBjxw5tmVqtxrZt29CnTx/JXufr16+xefNmANC2H5D1+9GkSRMsW7ZM2xMSERGBiRMnAgAGDRqEs2fPYuvWrbhx4wZ69OiBtm3b4t69e9mOiahAy7P7tBJRjg0YMEAYGRkJKysr7aN79+4Z1t22bZsoWbKk9nlAQICws7PTPrexsRGBgYEZrtuvXz8xbNgwnbLTp08LpVIpEhMTM1zn/e0/fvxYNGrUSJQtW1YkJyeLJk2aiKFDh+qs06NHD9G+fXshhBA//PCDqFy5skhJSclw+66uruLHH3/UPgcgdu3apVPn/dvLjx07VrRq1Ur7/PDhw8LU1FS8evUqV68TgLCyshKWlpbaW2F7eXllWD/Nh94PIYT4999/hUKhEE+fPtUp/+yzz8S3336b5faJCgtjedMbIspMy5Yt4e/vr31uZWUFADh+/Di+++47hIaGIjY2FiqVCklJSYiPj9fWeZevry98fHywceNGtG7dGj169ECFChUAACEhIfj333+xadMmbX0hBDQaDcLCwlCtWrUMY4uJiYG1tTWEEEhISMDHH3+MnTt3wtTUFLdv39YZpAkATZs2xfLlywEAPXr0wLJly1C+fHm0bdsW7du3R8eOHWFsnPOPoz59+qBx48Z49uwZnJ2dsWnTJrRv3x7FixfP1eu0sbHB1atXoVKpcPLkSSxevBirVq3SqaPv+wEAV69ehRAClStX1ilPTk7Ol7ElRPmBCQZRAWVlZYWKFSvqlD169Ajt27fHiBEjMHfuXJQoUQJnzpzBkCFDkJqamuF2Zs2ahd69e+OPP/7AwYMHMXPmTGzduhVdunSBRqPB8OHDdcZApClXrlymsaV98SqVSjg5OaX7IlUoFDrPhRDaMhcXF9y5cwfBwcH4888/MWrUKCxevBgnT57UOfWgjwYNGqBChQrYunUrRo4ciV27diEgIEC7PKevU6lUat+DqlWr4vnz5/D29sapU6cA5Oz9SIvHyMgIISEhMDIy0llmbW2t12snKqiYYBAVIleuXIFKpcIPP/wApfLtEKpt27Z9cL3KlSujcuXKmDBhAr788ksEBASgS5cu+Pjjj3Hr1q10icyHvPvF+75q1arhzJkz6N+/v7bs3LlzOr0EFhYW8PLygpeXF0aPHo2qVavi5s2b+Pjjj9Ntz8TEJFtXp/Tu3RubNm1C2bJloVQq0aFDB+2ynL7O902YMAFLly7Frl270KVLl2y9H6ampunir1u3LtRqNSIjI9GsWbNcxURUUHGQJ1EhUqFCBahUKvz000948OABNm7cmK7L/l2JiYkYM2YMTpw4gUePHuHs2bO4fPmy9sv+m2++wfnz5zF69Ghcv34d9+7dw969e/HVV1/lOMZJkyYhMDAQq1atwr1797B06VLs3LlTO7gxMDAQ69atw99//619DRYWFnB1dc1we25ubjh69CieP3+O6OjoTPfbp08fXL16FfPnz0f37t1hbm6uXSbV67S1tYWPjw9mzpwJIUS23g83Nze8efMGR48eRVRUFBISElC5cmX06dMH/fv3x86dOxEWFobLly9j4cKFOHDggF4xERVYcg4AIaKMDRgwQHTq1CnDZUuXLhWlS5cWFhYWwtPTU2zYsEEAENHR0UII3UGFycnJolevXsLFxUWYmpoKZ2dnMWbMGJ2BjZcuXRKff/65sLa2FlZWVqJ27dpi/vz5mcaW0aDF9/n5+Yny5csLExMTUblyZbFhwwbtsl27domGDRsKW1tbYWVlJRo1aiT+/PNP7fL3B3nu3btXVKxYURgbGwtXV1chRPpBnmnq168vAIhjx46lWybV63z06JEwNjYWQUFBQogPvx9CCDFixAhRsmRJAUDMnDlTCCFESkqKmDFjhnBzcxMmJiaiVKlSokuXLuLGjRuZxkRUmCiEEELeFIeIiIiKGp4iISIiIskxwSAiIiLJMcEgIiIiyTHBICIiIskxwSAiIiLJMcEgIiIiyTHBICIiIskxwSAiIiLJMcEgIiIiyTHBICIiIskxwSAiIiLJ/T9Q/LQa+N5OWgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 600x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr, tpr, thresholds=roc_curve(y_val, y_proba_ab)\n",
    "roc_auc=auc(fpr, tpr)\n",
    "\n",
    "plt.figure(figsize=(6,5))\n",
    "plt.plot(fpr, tpr, label=f'Logistic Regression (AUC = {roc_auc:.3f})')\n",
    "plt.plot([0, 1], [0, 1], 'k--', label='Random Classifier')\n",
    "\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curve')\n",
    "plt.legend(loc='lower right')\n",
    "plt.grid(alpha=0.3)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9fdc0052-4eb5-471e-99ed-fca4de9643f1",
   "metadata": {},
   "source": [
    "### 3th LGBMClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "4e99669e-9081-40c4-9f2f-248c31cf814c",
   "metadata": {},
   "outputs": [],
   "source": [
    "lg=LGBMClassifier(n_estimators=3000,learning_rate=0.03,num_leaves=31,max_depth=-1,min_child_samples=30,subsample=0.8,colsample_bytree=0.8,reg_lambda=1.0,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "20c7ff73-fd53-4839-8060-1c83d7135afe",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines\n",
      "[LightGBM] [Info] Number of positive: 160, number of negative: 1181\n",
      "[LightGBM] [Info] Auto-choosing row-wise multi-threading, the overhead of testing was 0.000698 seconds.\n",
      "You can set `force_row_wise=true` to remove the overhead.\n",
      "And if memory is not enough, you can set `force_col_wise=true`.\n",
      "[LightGBM] [Info] Total Bins 1592\n",
      "[LightGBM] [Info] Number of data points in the train set: 1341, number of used features: 47\n",
      "[LightGBM] [Info] [binary:BoostFromScore]: pavg=0.119314 -> initscore=-1.998943\n",
      "[LightGBM] [Info] Start training from score -1.998943\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n",
      "[LightGBM] [Warning] No further splits with positive gain, best gain: -inf\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-3 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: #000;\n",
       "  --sklearn-color-text-muted: #666;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\n",
       "}\n",
       "\n",
       "#sk-container-id-3 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-3 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-3 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-3 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-3 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: flex;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "  align-items: start;\n",
       "  justify-content: space-between;\n",
       "  gap: 0.5em;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 label.sk-toggleable__label .caption {\n",
       "  font-size: 0.6rem;\n",
       "  font-weight: lighter;\n",
       "  color: var(--sklearn-color-text-muted);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-3 div.sk-toggleable__content {\n",
       "  display: none;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  overflow: visible;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-3 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-3 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-3 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  display: inline-block;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-3 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-3 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-3 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 0.5em;\n",
       "  text-align: center;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-3 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-3 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-3 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".estimator-table summary {\n",
       "    padding: .5rem;\n",
       "    font-family: monospace;\n",
       "    cursor: pointer;\n",
       "}\n",
       "\n",
       ".estimator-table details[open] {\n",
       "    padding-left: 0.1rem;\n",
       "    padding-right: 0.1rem;\n",
       "    padding-bottom: 0.3rem;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table {\n",
       "    margin-left: auto !important;\n",
       "    margin-right: auto !important;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:nth-child(odd) {\n",
       "    background-color: #fff;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:nth-child(even) {\n",
       "    background-color: #f6f6f6;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:hover {\n",
       "    background-color: #e0e0e0;\n",
       "}\n",
       "\n",
       ".estimator-table table td {\n",
       "    border: 1px solid rgba(106, 105, 104, 0.232);\n",
       "}\n",
       "\n",
       ".user-set td {\n",
       "    color:rgb(255, 94, 0);\n",
       "    text-align: left;\n",
       "}\n",
       "\n",
       ".user-set td.value pre {\n",
       "    color:rgb(255, 94, 0) !important;\n",
       "    background-color: transparent !important;\n",
       "}\n",
       "\n",
       ".default td {\n",
       "    color: black;\n",
       "    text-align: left;\n",
       "}\n",
       "\n",
       ".user-set td i,\n",
       ".default td i {\n",
       "    color: black;\n",
       "}\n",
       "\n",
       ".copy-paste-icon {\n",
       "    background-image: url(data:image/svg+xml;base64,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);\n",
       "    background-repeat: no-repeat;\n",
       "    background-size: 14px 14px;\n",
       "    background-position: 0;\n",
       "    display: inline-block;\n",
       "    width: 14px;\n",
       "    height: 14px;\n",
       "    cursor: pointer;\n",
       "}\n",
       "</style><body><div id=\"sk-container-id-3\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>LGBMClassifier(colsample_bytree=0.8, learning_rate=0.03, min_child_samples=30,\n",
       "               n_estimators=3000, random_state=42, reg_lambda=1.0,\n",
       "               subsample=0.8)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" checked><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>LGBMClassifier</div></div><div><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></div></label><div class=\"sk-toggleable__content fitted\" data-param-prefix=\"\">\n",
       "        <div class=\"estimator-table\">\n",
       "            <details>\n",
       "                <summary>Parameters</summary>\n",
       "                <table class=\"parameters-table\">\n",
       "                  <tbody>\n",
       "                    \n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('boosting_type',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">boosting_type&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;gbdt&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('num_leaves',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">num_leaves&nbsp;</td>\n",
       "            <td class=\"value\">31</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_depth',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_depth&nbsp;</td>\n",
       "            <td class=\"value\">-1</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('learning_rate',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">learning_rate&nbsp;</td>\n",
       "            <td class=\"value\">0.03</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_estimators',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">n_estimators&nbsp;</td>\n",
       "            <td class=\"value\">3000</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('subsample_for_bin',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">subsample_for_bin&nbsp;</td>\n",
       "            <td class=\"value\">200000</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('objective',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">objective&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('class_weight',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">class_weight&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_split_gain',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_split_gain&nbsp;</td>\n",
       "            <td class=\"value\">0.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_child_weight',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_child_weight&nbsp;</td>\n",
       "            <td class=\"value\">0.001</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_child_samples',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_child_samples&nbsp;</td>\n",
       "            <td class=\"value\">30</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('subsample',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">subsample&nbsp;</td>\n",
       "            <td class=\"value\">0.8</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('subsample_freq',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">subsample_freq&nbsp;</td>\n",
       "            <td class=\"value\">0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('colsample_bytree',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">colsample_bytree&nbsp;</td>\n",
       "            <td class=\"value\">0.8</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('reg_alpha',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">reg_alpha&nbsp;</td>\n",
       "            <td class=\"value\">0.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('reg_lambda',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">reg_lambda&nbsp;</td>\n",
       "            <td class=\"value\">1.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('random_state',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">random_state&nbsp;</td>\n",
       "            <td class=\"value\">42</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_jobs',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">n_jobs&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('importance_type',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">importance_type&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;split&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "                  </tbody>\n",
       "                </table>\n",
       "            </details>\n",
       "        </div>\n",
       "    </div></div></div></div></div><script>function copyToClipboard(text, element) {\n",
       "    // Get the parameter prefix from the closest toggleable content\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${text}` : text;\n",
       "\n",
       "    const originalStyle = element.style;\n",
       "    const computedStyle = window.getComputedStyle(element);\n",
       "    const originalWidth = computedStyle.width;\n",
       "    const originalHTML = element.innerHTML.replace('Copied!', '');\n",
       "\n",
       "    navigator.clipboard.writeText(fullParamName)\n",
       "        .then(() => {\n",
       "            element.style.width = originalWidth;\n",
       "            element.style.color = 'green';\n",
       "            element.innerHTML = \"Copied!\";\n",
       "\n",
       "            setTimeout(() => {\n",
       "                element.innerHTML = originalHTML;\n",
       "                element.style = originalStyle;\n",
       "            }, 2000);\n",
       "        })\n",
       "        .catch(err => {\n",
       "            console.error('Failed to copy:', err);\n",
       "            element.style.color = 'red';\n",
       "            element.innerHTML = \"Failed!\";\n",
       "            setTimeout(() => {\n",
       "                element.innerHTML = originalHTML;\n",
       "                element.style = originalStyle;\n",
       "            }, 2000);\n",
       "        });\n",
       "    return false;\n",
       "}\n",
       "\n",
       "document.querySelectorAll('.fa-regular.fa-copy').forEach(function(element) {\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const paramName = element.parentElement.nextElementSibling.textContent.trim();\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${paramName}` : paramName;\n",
       "\n",
       "    element.setAttribute('title', fullParamName);\n",
       "});\n",
       "</script></body>"
      ],
      "text/plain": [
       "LGBMClassifier(colsample_bytree=0.8, learning_rate=0.03, min_child_samples=30,\n",
       "               n_estimators=3000, random_state=42, reg_lambda=1.0,\n",
       "               subsample=0.8)"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lg.fit(x_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "a7f489f5-afcc-4317-a58c-cb7c95f6380e",
   "metadata": {},
   "outputs": [],
   "source": [
    "lgtahmin=lg.predict(x_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "e63e5a26-a189-48e1-b1ee-fc5e042302ee",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_proba_lg=lg.predict_proba(x_val)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "eba02bbf-1bc8-485e-8254-a67e45644234",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC: 0.8203\n"
     ]
    }
   ],
   "source": [
    "roc_auc=roc_auc_score(y_val, y_proba_lg)\n",
    "print(f\"ROC AUC: {roc_auc:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "b434ad22-1fbb-41b2-aad0-8475a7fec65f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.8869\n"
     ]
    }
   ],
   "source": [
    "acc=accuracy_score(y_val, lgtahmin)\n",
    "print(f\"Accuracy: {acc:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "9e7c2f7e-8094-450f-852c-08874318f104",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[290,   6],\n",
       "       [ 32,   8]])"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(y_val, lgtahmin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "00777052-fbab-4823-90db-476701e93d89",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.90      0.98      0.94       296\n",
      "         1.0       0.57      0.20      0.30        40\n",
      "\n",
      "    accuracy                           0.89       336\n",
      "   macro avg       0.74      0.59      0.62       336\n",
      "weighted avg       0.86      0.89      0.86       336\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"\\nClassification Report:\")\n",
    "print(classification_report(y_val, lgtahmin))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "8d64f31c-1e5f-400b-b287-e6fc340c7c3c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr, tpr, thresholds=roc_curve(y_val, y_proba_lg)\n",
    "roc_auc=auc(fpr, tpr)\n",
    "\n",
    "plt.figure(figsize=(6,5))\n",
    "plt.plot(fpr, tpr, label=f'Logistic Regression (AUC = {roc_auc:.3f})')\n",
    "plt.plot([0, 1], [0, 1], 'k--', label='Random Classifier')\n",
    "\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curve')\n",
    "plt.legend(loc='lower right')\n",
    "plt.grid(alpha=0.3)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca2207d9-7003-482e-a82f-6eab4f87187e",
   "metadata": {},
   "source": [
    "### 4th XGBClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "f3e37262-517e-4b54-bbf3-a3b340790890",
   "metadata": {},
   "outputs": [],
   "source": [
    "xg=XGBClassifier(objective=\"binary:logistic\",n_estimators=4000,learning_rate=0.03,max_depth=4,min_child_weight=5,subsample=0.8,colsample_bytree=0.8,gamma=0.0,reg_lambda=1.0,reg_alpha=0.0,eval_metric=\"auc\",random_state=42,n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "id": "588f6798-7a2d-45dc-ac19-53126d12c675",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-6 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: #000;\n",
       "  --sklearn-color-text-muted: #666;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\n",
       "}\n",
       "\n",
       "#sk-container-id-6 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-6 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-6 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-6 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-6 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: flex;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "  align-items: start;\n",
       "  justify-content: space-between;\n",
       "  gap: 0.5em;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 label.sk-toggleable__label .caption {\n",
       "  font-size: 0.6rem;\n",
       "  font-weight: lighter;\n",
       "  color: var(--sklearn-color-text-muted);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-6 div.sk-toggleable__content {\n",
       "  display: none;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  overflow: visible;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-6 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-6 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-6 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-6 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-6 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-6 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  display: inline-block;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-6 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-6 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-6 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 0.5em;\n",
       "  text-align: center;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-6 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-6 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-6 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".estimator-table summary {\n",
       "    padding: .5rem;\n",
       "    font-family: monospace;\n",
       "    cursor: pointer;\n",
       "}\n",
       "\n",
       ".estimator-table details[open] {\n",
       "    padding-left: 0.1rem;\n",
       "    padding-right: 0.1rem;\n",
       "    padding-bottom: 0.3rem;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table {\n",
       "    margin-left: auto !important;\n",
       "    margin-right: auto !important;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:nth-child(odd) {\n",
       "    background-color: #fff;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:nth-child(even) {\n",
       "    background-color: #f6f6f6;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:hover {\n",
       "    background-color: #e0e0e0;\n",
       "}\n",
       "\n",
       ".estimator-table table td {\n",
       "    border: 1px solid rgba(106, 105, 104, 0.232);\n",
       "}\n",
       "\n",
       ".user-set td {\n",
       "    color:rgb(255, 94, 0);\n",
       "    text-align: left;\n",
       "}\n",
       "\n",
       ".user-set td.value pre {\n",
       "    color:rgb(255, 94, 0) !important;\n",
       "    background-color: transparent !important;\n",
       "}\n",
       "\n",
       ".default td {\n",
       "    color: black;\n",
       "    text-align: left;\n",
       "}\n",
       "\n",
       ".user-set td i,\n",
       ".default td i {\n",
       "    color: black;\n",
       "}\n",
       "\n",
       ".copy-paste-icon {\n",
       "    background-image: url(data:image/svg+xml;base64,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);\n",
       "    background-repeat: no-repeat;\n",
       "    background-size: 14px 14px;\n",
       "    background-position: 0;\n",
       "    display: inline-block;\n",
       "    width: 14px;\n",
       "    height: 14px;\n",
       "    cursor: pointer;\n",
       "}\n",
       "</style><body><div id=\"sk-container-id-6\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=0.8, device=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=&#x27;auc&#x27;, feature_types=None,\n",
       "              feature_weights=None, gamma=0.0, grow_policy=None,\n",
       "              importance_type=None, interaction_constraints=None,\n",
       "              learning_rate=0.03, max_bin=None, max_cat_threshold=None,\n",
       "              max_cat_to_onehot=None, max_delta_step=None, max_depth=4,\n",
       "              max_leaves=None, min_child_weight=5, missing=nan,\n",
       "              monotone_constraints=None, multi_strategy=None, n_estimators=4000,\n",
       "              n_jobs=-1, num_parallel_tree=None, ...)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-8\" type=\"checkbox\" checked><label for=\"sk-estimator-id-8\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>XGBClassifier</div></div><div><a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.1.0/python/python_api.html#xgboost.XGBClassifier\">?<span>Documentation for XGBClassifier</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></div></label><div class=\"sk-toggleable__content fitted\" data-param-prefix=\"\">\n",
       "        <div class=\"estimator-table\">\n",
       "            <details>\n",
       "                <summary>Parameters</summary>\n",
       "                <table class=\"parameters-table\">\n",
       "                  <tbody>\n",
       "                    \n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('objective',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">objective&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;binary:logistic&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('base_score',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">base_score&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('booster',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">booster&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('callbacks',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">callbacks&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('colsample_bylevel',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">colsample_bylevel&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('colsample_bynode',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">colsample_bynode&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('colsample_bytree',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">colsample_bytree&nbsp;</td>\n",
       "            <td class=\"value\">0.8</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('device',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">device&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('early_stopping_rounds',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">early_stopping_rounds&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('enable_categorical',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">enable_categorical&nbsp;</td>\n",
       "            <td class=\"value\">False</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('eval_metric',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">eval_metric&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;auc&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('feature_types',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">feature_types&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('feature_weights',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">feature_weights&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('gamma',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">gamma&nbsp;</td>\n",
       "            <td class=\"value\">0.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('grow_policy',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">grow_policy&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('importance_type',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">importance_type&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('interaction_constraints',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">interaction_constraints&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('learning_rate',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">learning_rate&nbsp;</td>\n",
       "            <td class=\"value\">0.03</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_bin',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_bin&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_cat_threshold',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_cat_threshold&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_cat_to_onehot',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_cat_to_onehot&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_delta_step',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_delta_step&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_depth',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_depth&nbsp;</td>\n",
       "            <td class=\"value\">4</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_leaves',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_leaves&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_child_weight',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_child_weight&nbsp;</td>\n",
       "            <td class=\"value\">5</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('missing',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">missing&nbsp;</td>\n",
       "            <td class=\"value\">nan</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('monotone_constraints',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">monotone_constraints&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('multi_strategy',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">multi_strategy&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_estimators',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">n_estimators&nbsp;</td>\n",
       "            <td class=\"value\">4000</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_jobs',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">n_jobs&nbsp;</td>\n",
       "            <td class=\"value\">-1</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('num_parallel_tree',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">num_parallel_tree&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('random_state',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">random_state&nbsp;</td>\n",
       "            <td class=\"value\">42</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('reg_alpha',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">reg_alpha&nbsp;</td>\n",
       "            <td class=\"value\">0.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('reg_lambda',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">reg_lambda&nbsp;</td>\n",
       "            <td class=\"value\">1.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('sampling_method',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">sampling_method&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('scale_pos_weight',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">scale_pos_weight&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('subsample',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">subsample&nbsp;</td>\n",
       "            <td class=\"value\">0.8</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('tree_method',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">tree_method&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('validate_parameters',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">validate_parameters&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('verbosity',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">verbosity&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "                  </tbody>\n",
       "                </table>\n",
       "            </details>\n",
       "        </div>\n",
       "    </div></div></div></div></div><script>function copyToClipboard(text, element) {\n",
       "    // Get the parameter prefix from the closest toggleable content\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${text}` : text;\n",
       "\n",
       "    const originalStyle = element.style;\n",
       "    const computedStyle = window.getComputedStyle(element);\n",
       "    const originalWidth = computedStyle.width;\n",
       "    const originalHTML = element.innerHTML.replace('Copied!', '');\n",
       "\n",
       "    navigator.clipboard.writeText(fullParamName)\n",
       "        .then(() => {\n",
       "            element.style.width = originalWidth;\n",
       "            element.style.color = 'green';\n",
       "            element.innerHTML = \"Copied!\";\n",
       "\n",
       "            setTimeout(() => {\n",
       "                element.innerHTML = originalHTML;\n",
       "                element.style = originalStyle;\n",
       "            }, 2000);\n",
       "        })\n",
       "        .catch(err => {\n",
       "            console.error('Failed to copy:', err);\n",
       "            element.style.color = 'red';\n",
       "            element.innerHTML = \"Failed!\";\n",
       "            setTimeout(() => {\n",
       "                element.innerHTML = originalHTML;\n",
       "                element.style = originalStyle;\n",
       "            }, 2000);\n",
       "        });\n",
       "    return false;\n",
       "}\n",
       "\n",
       "document.querySelectorAll('.fa-regular.fa-copy').forEach(function(element) {\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const paramName = element.parentElement.nextElementSibling.textContent.trim();\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${paramName}` : paramName;\n",
       "\n",
       "    element.setAttribute('title', fullParamName);\n",
       "});\n",
       "</script></body>"
      ],
      "text/plain": [
       "XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=0.8, device=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric='auc', feature_types=None,\n",
       "              feature_weights=None, gamma=0.0, grow_policy=None,\n",
       "              importance_type=None, interaction_constraints=None,\n",
       "              learning_rate=0.03, max_bin=None, max_cat_threshold=None,\n",
       "              max_cat_to_onehot=None, max_delta_step=None, max_depth=4,\n",
       "              max_leaves=None, min_child_weight=5, missing=nan,\n",
       "              monotone_constraints=None, multi_strategy=None, n_estimators=4000,\n",
       "              n_jobs=-1, num_parallel_tree=None, ...)"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xg.fit(x_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "id": "ae2bca3c-99fe-4374-adc0-724f63a27a75",
   "metadata": {},
   "outputs": [],
   "source": [
    "xgtahmin=xg.predict(x_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "id": "a1e91b39-185c-4550-8ca3-6900e4a54f08",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_proba_xg=xg.predict_proba(x_val)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "id": "eddb91f7-9efb-4ab0-ba37-04dd8baf004c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC: 0.8227\n"
     ]
    }
   ],
   "source": [
    "roc_auc=roc_auc_score(y_val, y_proba_xg)\n",
    "print(f\"ROC AUC: {roc_auc:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "id": "12a3952d-3b54-4663-8a2f-8b3ef398c05a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.8810\n"
     ]
    }
   ],
   "source": [
    "acc=accuracy_score(y_val, xgtahmin)\n",
    "print(f\"Accuracy: {acc:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "id": "dab818a3-e559-4409-ab3f-d71c75ef6dcc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[287,   9],\n",
       "       [ 31,   9]])"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(y_val, xgtahmin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "id": "db3b1f9d-9190-4e86-bdb4-322766ccefe7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.90      0.97      0.93       296\n",
      "         1.0       0.50      0.23      0.31        40\n",
      "\n",
      "    accuracy                           0.88       336\n",
      "   macro avg       0.70      0.60      0.62       336\n",
      "weighted avg       0.85      0.88      0.86       336\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"\\nClassification Report:\")\n",
    "print(classification_report(y_val, xgtahmin))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "id": "6cce6015-b8df-434f-bd2b-f01f20b18d4f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhgAAAHUCAYAAAB4RlFCAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAfDtJREFUeJzt3Xl8TNf/P/DXTPZdkERCJLEvpVQsoapUxVKxS2tfYldLiiofu1YtVbREbQlqrZ2iYt+JWCuKEmKJRogkss/M+f3hm/kZWWSSO7lZXs/HI4+HOffcO+/7njHznnvPPVchhBAgIiIikpBS7gCIiIio6GGBQURERJJjgUFERESSY4FBREREkmOBQURERJJjgUFERESSY4FBREREkmOBQURERJJjgUFERESSY4FBVAwEBQVBoVBo/4yNjeHs7Iwvv/wSd+/ezXSdtLQ0BAQEwMvLC3Z2drCwsED16tUxceJEvHjxItN1NBoN1q9fj5YtW6J06dIwMTGBo6MjvvjiC+zduxcajea9saakpODXX3/Fxx9/DHt7e5iamqJs2bLo3r07Tpw4kac8EFH+YYFBVIwEBgbi3LlzOHz4MEaOHIk9e/bg448/RkxMjE6/xMREfP755/j6669Rt25dbNq0Cfv370fv3r2xYsUK1K1bF7dv39ZZJzk5GW3btkXfvn3h6OiIgIAAHD16FMuXL4eLiwu6deuGvXv3ZhtfdHQ0mjRpAn9/f3zwwQcICgrCkSNH8NNPP8HIyAifffYZrl27JnleiMgABBEVeYGBgQKACAkJ0WmfMWOGACDWrFmj0z548GABQGzevDnDtm7fvi3s7OxEzZo1hUql0rYPGzZMABBr167NNIY7d+6Ia9euZRtnmzZthLGxsThy5Eimyy9evCgePnyY7TZyKjExUZLtEFHmeASDqBjz9PQEAPz333/atmfPnmHNmjXw9vaGr69vhnWqVKmCb7/9Fjdv3sSuXbu066xatQre3t7o06dPps9VuXJl1K5dO8tYQkNDceDAAQwcOBAtWrTItE/9+vVRvnx5AMD06dOhUCgy9Ek/HfTgwQNtm7u7O7744gvs2LEDdevWhbm5OWbMmIG6deuiadOmGbahVqtRtmxZdO7cWduWmpqK2bNno1q1ajAzM4ODgwP69++P58+fZ7lPRMUZCwyiYiw8PBzAm6Ih3bFjx6BSqdCxY8cs10tfFhwcrF0nLS0t23Xe59ChQzrbltrly5cxfvx4jBo1CgcPHkSXLl3Qv39/nD59OsM4lEOHDuHp06fo378/gDdjSzp06IAff/wRPXr0wJ9//okff/wRwcHB+PTTT5GUlGSQmIkKM2O5AyCi/KNWq6FSqZCcnIwzZ85g9uzZ+OSTT+Dj46PtExERAQDw8PDIcjvpy9L75mSd95FiG9mJiopCWFiYTjFVoUIFjB8/HkFBQfj++++17UFBQXByckKbNm0AAFu3bsXBgwexfft2naMaH374IerXr4+goCAMGzbMIHETFVY8gkFUjDRq1AgmJiawsbFB69atYW9vj927d8PYOHe/NTI7RVFQ1a5dW6e4AIBSpUqhffv2WLt2rfYKl5iYGOzevRt9+vTR5mXfvn0oUaIE2rdvD5VKpf2rU6cOypQpg+PHj+f37hAVeCwwiIqRdevWISQkBEePHsWQIUNw69YtfPXVVzp90sc4pJ8+yUz6MldX1xyv8z5SbCM7zs7OmbYPGDAAT5480Z7u2bRpE1JSUtCvXz9tn//++w+vXr2CqakpTExMdP6ePXuG6Ohog8RMVJixwCAqRqpXrw5PT080b94cy5cvh5+fHw4ePIht27Zp+zRv3hzGxsbaAZyZSV/2+eefa9cxMTHJdp338fb21tn2+5ibmwN4M2/G27L6ss/qaIu3tzdcXFwQGBgI4M2lvA0bNkSNGjW0fUqXLo1SpUohJCQk079ly5blKGaiYkXuy1iIyPCyukz15cuXwt7eXlSvXl2o1WptuyEuU/3333/zfJlqSEiI9jLVTZs2CQDi4sWLOn0++eQTAUCEh4dr29zc3ES7du2yfN5vv/1WmJmZiZMnTwoA4rffftNZ/vvvvwsA4vz589nGT0T/HwsMomIgqwJDCCHmzZsnAIj169dr216/fi2aNWsmjI2NxfDhw8WBAwfE0aNHxQ8//CBKliwpypUrJ/755x+d7SQlJQlvb2+hUChEjx49xB9//CFOnjwpduzYIYYNGybMzc3Frl27so3z+fPnol69esLU1FQMHTpU7N69W5w8eVJs2bJF9OrVSxgZGYmrV68KIYSIjY0VJUuWFLVq1RI7d+4Ue/fuFV26dBEeHh56Fxi3b98WAES5cuWEhYWFePXqlc5ylUol2rRpI0qWLClmzJghDhw4IA4fPiyCgoJE3759xY4dO7LdL6LiiAUGUTGQXYGRlJQkypcvLypXrqxzRCI1NVUsXbpUNGzYUFhbWwszMzNRtWpVMWHCBBEdHZ3p86hUKrF27VrRokULUbJkSWFsbCwcHBxEmzZtxMaNG3WOkmQlKSlJLFmyRHh5eQlbW1thbGwsXFxcROfOncWff/6p0/fixYuicePGwsrKSpQtW1ZMmzZNrFq1Su8CQwghGjduLACInj17Zro8LS1NLFiwQHz44YfC3NxcWFtbi2rVqokhQ4aIu3fvvne/iIobhRBCyHFqhoiIiIouDvIkIiIiybHAICIiIsmxwCAiIiLJscAgIiIiybHAICIiIsmxwCAiIiLJFbu7qWo0Gjx9+hQ2NjaF6kZNREREchNCID4+Hi4uLlAqsz9GUewKjKdPn2pv0ERERET6e/ToEcqVK5dtn2JXYNjY2AB4kxxbW1vJtqvRaPD8+XM4ODi8t6qj92M+pcecSov5lB5zKi1D5DMuLg6urq7a79LsFLsCI/20iK2treQFRnJyMmxtbfkfQwLMp/SYU2kxn9JjTqVlyHzmZIgBX0EiIiKSHAsMIiIikhwLDCIiIpIcCwwiIiKSHAsMIiIikhwLDCIiIpIcCwwiIiKSHAsMIiIikhwLDCIiIpIcCwwiIiKSnKwFxsmTJ9G+fXu4uLhAoVBg165d713nxIkTqFevHszNzVGhQgUsX77c8IESERGRXmQtMBISEvDhhx/i119/zVH/8PBwtG3bFk2bNsWVK1cwadIkjBo1Ctu3bzdwpERERKQPWW921qZNG7Rp0ybH/ZcvX47y5ctj0aJFAIDq1avj0qVLWLBgAbp06WKgKImIqDAQQiApTS13GAWGRqNBUpoaQghZnr9Q3U313LlzaNWqlU6bt7c3Vq9ejbS0NJiYmGRYJyUlBSkpKdrHcXFxAN4kXqPRSBabRqOBEELSbRZnzKf0mFNpMZ/Sy0tOhRDo/tt5hEa8kj6wQkgIDZLuh8KyYn1cn/oZrM3ff/fTnNDntSlUBcazZ8/g5OSk0+bk5ASVSoXo6Gg4OztnWGfOnDmYMWNGhvbnz58jOTlZstg0Gg1iY2MhhOBthiXAfEqPOZUW8ym9vOQ0KU3N4uL/aFKT8WL/z0i8fQYlWw3H8+e1kWiW8Qd4bsTHx+e4b6EqMICM96BPP/ST1b3pv/vuO/j7+2sfx8XFwdXVFQ4ODrC1tZUsLo1GA4VCAQcHB37YSID5lB5zKi3mU3p5yWliqkr774uTWsDS1Ejq8AqFx48e4cvuXfHo9lWYmJjg288roLxLGRgZSZMPc3PzHPctVAVGmTJl8OzZM522qKgoGBsbo1SpUpmuY2ZmBjMzswztSqVS8g8FhUJhkO0WV8yn9JhTaTGf0sttTt/ub21uAkvTQvX1Jonz58+jY8eO+O+//+Dg4IBt27ahSpUqMDIykuw9qs92CtX/Ci8vLwQHB+u0HTp0CJ6enpmOvyAiIioO1q9fj08//RT//fcfateujZCQEHz88ceyxiRrgfH69WtcvXoVV69eBfDmMtSrV68iIiICwJvTG3369NH2Hzp0KB4+fAh/f3/cunULa9aswerVqzFu3Dg5wiciIpLdP//8g759+yIlJQUdO3bEmTNn4ObmJndY8p4iuXTpEpo3b659nD5Wom/fvggKCkJkZKS22AAADw8P7N+/H2PHjsXSpUvh4uKCJUuW8BJVIiIqtqpVq4a5c+ciNjYWM2fOLDCn7GQtMD799NNsr88NCgrK0NasWTNcvnzZgFEREREVbOHh4VAoFHB3dwcAjB8/Xt6AMlH8RsEQERkAJ3nKu/SJoRJTVbm4iqT45P7kyZPo3LkznJ2dcfbsWdjY2MgdUqZYYBAR5ZEQAl2Xn0Powxi5Q6EibuXKlRg+fDhUKhXc3d2RkJBQYAuMgnGihoioEEtKU7O4KCA83exhYVL05sBQqVQYPXo0Bg8eDJVKBV9fX5w8eRJlypSRO7Qs8QgGEZGELv2vZbGd5CmvNBoNnj9/nqfJyyxMjLKceLGwiomJga+vr3aahlmzZmHy5MkFfj9ZYBARScjS1KhYTvIkBY1GAwuTN/krKFdCFAQjRoxAcHAwLC0tsX79enTu3FnukHKE/wuIiIgKsAULFiA8PBwBAQGoU6eO3OHkGEtEIiKiAkQIgYsXL2ofu7i44OzZs4WquABYYBARERUYaWlpGDZsGBo2bIgtW7Zo2wv6eIvM8BQJERFRAfDixQt07doVx48fh0KhyHBzz8KGBQYREfSfKOvtSaGSVVnPSEyUEzdv3oSPjw/u378PGxsbbNq0Ce3atZM7rDxhgUFExR4nyiI57du3Dz169EB8fDwqVKiAPXv2oGbNmnKHlWcsMIio2JNqoqyiOskTGc6tW7fQoUMHaDQaNGvWDNu2bUPp0qXlDksSLDCIiN6S04myMpsUqihO8kSGVb16dXzzzTeIjY3FL7/8AlNTU7lDkgwLDCKit+R0oixOCkW59ezZMyiVSjg6OgIAfvzxRygUiiJXnPJ/BRERUT65cuUKGjRogM6dOyMlJQUAoFQqi1xxAbDAICIiyhfbt2/Hxx9/jEePHuH58+eIioqSOySDYoFBRERkQEIIzJo1C127dkViYiJatWqFCxcuwNXVVe7QDIpjMIioWBNCIDE15/NfEOkjMTERAwYM0M7KOXr0aCxYsADGxkX/67fo7yERURY4/wUZ2qBBg7BlyxaYmJhg2bJl8PPzkzukfMNTJERUbL07/wXnsSCpTZ8+HVWqVMHhw4eLVXEB8AgGERGAN/NflLIyLZKj+Sl//fvvv6hUqRIAoHLlyggLC4ORUfErXHkEg4gIb+a/YHFBeaHRaDBp0iRUq1YNwcHB2vbiWFwALDCIiIjyLD4+Hp07d8acOXOgVqtx4cIFuUOSHU+REBER5cGDBw/g4+ODGzduwMzMDKtWrUKvXr3kDkt2LDCIiIhy6dSpU+jcuTOio6Ph5OSEXbt2oVGjRnKHVSCwwCAiIsqFv//+G5999hnS0tJQt25d7N69u8hPnqUPFhhEVGwJIXcEVJjVrFkTvXr1wuvXrxEYGAgrKyu5QypQWGAQUbEkhEC35efkDoMKmdjYWCgUCtja2kKhUOC3336DkZER76ibCWaEiIqlpDQ1wiLjAAA1nG05wRa917///otGjRqhR48eUKvfTC9vYmLC4iILzAoRFXt/DPXiHBiUraNHj6JBgwb4559/cO3aNTx+/FjukAo8FhhEVOyxtqDsLFu2DK1atUJMTAwaNmyIixcvws3NTe6wCjwWGERERJlIS0vD8OHDMWLECKjVavTq1QvHjx+Hs7Oz3KEVCiwwiIiIMjFw4EAEBARAoVDgxx9/xLp162Bubi53WIUGCwwiIqJMjB49Go6Ojti1axe+/fZbjtPREy9TJSIi+j9RUVFwdHQEANSrVw/h4eGwtLSUOarCiQUGERmUEAJJaWq5w8ggMbXgxUTyEULg559/xpQpU3Ds2DE0aNAAAFhc5AELDCIyGCEEui4/h9CHMXKHQpSllJQUDBs2DIGBgQCAnTt3agsMyj0WGERkMElp6gJfXHi62XOSrWIsKioKnTt3xpkzZ6BUKrFw4UKMGjVK7rCKBBYYRJQvLv2vJSxNC94XuYWJEQfvFVPXr19H+/btERERATs7O2zZsgXe3t5yh1VksMAgonxhaWoES1N+5FDB8Pfff6Nx48ZISEhA5cqVsXfvXlStWlXusIoU/m8nIqJip0aNGmjZsiUSEhKwdetW2Nvbyx1SkcMCg4iIioWkpCQoFAqYm5tDqVRi48aNMDU1hbExvwoNgRNtERFRkff06VN88sknGDx4MIQQAN5cgsriwnCYWSIiKtJCQkLQsWNHPH36FPfv30dERARvVpYPWGAQySg/J6HSaDRISlMjMVUFpTJ/Dl5yMiuS2+bNm9G/f38kJyejRo0a2LNnD4uLfMICg0gmnISKyHA0Gg2mTZuG2bNnAwDatm2LTZs2wdbWVubIig+OwSCSSWGYhEoqnMyK8tuQIUO0xcW4ceOwZ88eFhf5jEcwiAqA/JiESqPR4Pnz53BwcMi3UyTpOJkV5bcePXpg06ZNWLp0Kfr27St3OMUSCwyiAiA/JqHSaDSwMHnzPPldYBDlh4SEBFhZWQEAmjdvjgcPHqB06dIyR1V88VOGiIgKvbVr16JChQr4559/tG0sLuTFAoOIiAottVqN8ePHo1+/foiKisJvv/0md0j0f3iKhIiICqW4uDh89dVX2L9/PwBgypQpmD59urxBkRYLDKJ89Pa8F5wjgij37t27Bx8fH4SFhcHc3BxBQUHw9fWVOyx6CwsMonzCeS+IpHHz5k188sknePnyJVxcXLB79254enrKHRa9gwUGUT7Jat4LzhFBpJ/KlSvjgw8+QFJSEnbt2gUXFxe5Q6JMsMAgksHb815wjgii91OpVAAAY2NjmJqaYufOnbCwsICFhYXMkVFWeBUJkQzS572wNDVmcUH0HjExMWjTpg0mTJigbStZsiSLiwKOBQYRERVY//zzDxo2bIjDhw9jxYoViIiIkDskyiEWGEREVCD99ddfaNSoEe7evQs3NzecPXsW5cuXlzssyiHZC4xly5bBw8MD5ubmqFevHk6dOpVt/w0bNuDDDz+EpaUlnJ2d0b9/f7x48SKfoiUiIkMTQmDx4sVo27YtYmNj8fHHH+PixYuoXbu23KGRHmQtMLZs2YIxY8Zg8uTJuHLlCpo2bYo2bdpkeQjs9OnT6NOnDwYOHIibN2/ijz/+QEhICPz8/PI5ciIiMhR/f3+MGTMGGo0G/fv3x+HDh+Ho6Ch3WKQnWQuMhQsXYuDAgfDz80P16tWxaNEiuLq6IiAgINP+58+fh7u7O0aNGgUPDw98/PHHGDJkCC5dupTPkVNxIIRAYqpKwj9OrEWUE02bNoWRkREWLlyI1atXw8zMTO6QKBdku0w1NTUVoaGhmDhxok57q1atcPbs2UzXady4MSZPnoz9+/ejTZs2iIqKwrZt29CuXbssnyclJQUpKSnax3FxcQDe3FlSo9FIsCfQbk8IIek2izO58ymEQPffziM04pVBti/1+y+nz8n3qHSYT2mpVCoolUoIIdCxY0fcvn0bHh4eEEJACCF3eIWSId6j+mxLtgIjOjoaarUaTk5OOu1OTk549uxZpus0btwYGzZsgK+vL5KTk6FSqeDj44Nffvkly+eZM2cOZsyYkaH9+fPnSE5OzttOvEWj0SA2NhZCCN4KWwJy5zMpTW2w4qK2ixXiY17gdT5fnip3Tosa5lM6f/31F2bMmIEtW7bA2toaQghYWVkhKipK7tAKNUO8R+Pj43PcV/aJtt6dA0AIkeW8AGFhYRg1ahSmTp0Kb29vREZGYvz48Rg6dChWr16d6Trfffcd/P39tY/j4uLg6uoKBwcH2NraSrYfGo0GCoUCDg4O/LCRgNz5TExVaf99cVIL7aRYUpBrYi25c1rUMJ95J4TAvHnzMHnyZAghEBQUhP/973/MqUQM8R41NzfPcV/ZCozSpUvDyMgow9GKqKioDEc10s2ZMwdNmjTB+PHjAQC1a9eGlZUVmjZtitmzZ8PZ2TnDOmZmZpmev1MqlZK/gRUKhUG2W1zJmc+3n9Pa3ASWprLX4pLge1RazGfuJScnw8/PDxs2bAAADBs2DAsXLkRMTAxzKiGp36P6bEe2V9DU1BT16tVDcHCwTntwcDAaN26c6TqJiYkZds7I6M0vS56jIyIqHCIjI/Hpp59iw4YNMDIywtKlS7Fs2TKYmJjIHRpJSNafZf7+/ujduzc8PT3h5eWlnaVt6NChAN6c3njy5AnWrVsHAGjfvj0GDRqEgIAA7SmSMWPGoEGDBrzZDRFRIXD79m20bNkSjx8/hr29PbZt24YWLVrIHRYZgKwFhq+vL168eIGZM2ciMjISH3zwAfbv3w83NzcAb6rct+fE6NevH+Lj4/Hrr7/im2++QYkSJdCiRQvMnTtXrl0gIiI9lCtXDqVKlYK1tTX27t2LSpUqyR0SGYhCFLNzC3FxcbCzs0NsbKzkgzyjoqLg6OjIc4cSkDufiakq1Jj6FwAgbKZ3kRiDIXdOixrmM+fSBxumD25+8uQJrK2tYWdnl6EfcyodQ+RTn+9QvoJEyGxSLU6KRSSFhIQE+Pr6Ys6cOdq2smXLZiguqOgp/D/LiPJICIGuy88h9GGM3KEQFSmPHj1Chw4dcOXKFezduxd9+vRBuXLl5A6L8gmPYFCxl5SmzrK48HSzh4WJdHNgEBUX58+fR/369XHlyhU4ODjg8OHDLC6KGR7BIHrLpf+11JlUS65JsYgKs/Xr12PQoEFISUlBrVq1sGfPHri7u8sdFuUzHsEgeoulqREsTY21fywuiPQzZcoU9OnTBykpKejQoQPOnj3L4qKYYoFBRESSSZ9mYNKkSdixYwesra1ljojkwlMkRESUJ2/fQ8rPzw9169ZFvXr1ZI6K5MYjGERElGsnTpxAo0aNEB0drW1jcUEACwwiIsqllStXomXLlrh48SJmzJghdzhUwPAUCeUrIQSS0t4/iZVGo0FSmhqJqSqDz+jHSbWI9KNSqfDNN99gyZIlAN7c9oG3bKB3scCgfMMJrYgKv5iYGPj6+mrvhD1r1ixMnjyZV1xRBiwwKN9kN6FVQcBJtYiyd//+fbRp0wZ37tyBpaUl1q9fj86dO8sdFhVQLDBIFu9OaPUujUaD58+fw8HBId9uesRJtYiyZ2dnB7VaDVdXV+zZswd16tSROyQqwFhgkCzSJ7TKikajgYXJmz68qyJRwVCqVCns378fdnZ2cHJykjscKuD4yU1ERJlKTU3FkCFDsGLFCm1blSpVWFxQjvAIBhERZRAdHY2uXbvixIkTMDMzwxdffAEXFxe5w6JChAUGERHpuHnzJtq3b4/w8HDY2Nhg48aNLC5IbzxFQkREWvv27YOXlxfCw8NRoUIFnDt3Dl988YXcYVEhxAKDiIgAAAsWLICPjw/i4+PRrFkzXLhwATVr1pQ7LCqkWGAQERGANzN0CiEwZMgQHDp0CKVLl5Y7JCrEOAaDiIgAAN9++y3q1q2LVq1acU4YyjMewSAiKqauXr2K9u3b4/Xr1wAAhUIBb29vFhckCRYYRETF0I4dO9CkSRPs27cPkydPljscKoJYYBARFSNCCMyaNQtdunRBYmIiWrVqxVutk0FwDAYRUTGRmJiIAQMGYMuWLQCA0aNHY8GCBTA25lcBSY/vKiKiYuDJkyfo0KEDQkNDYWxsjGXLlmHQoEFyh0VFGAsMkowQAklp6iyXJ6ZmvYyIDEsIgSdPnqBUqVLYvn07mjVrJndIVMSxwCBJCCHQdfk5hD6MkTsUIspEuXLlsG/fPpQsWRIeHh5yh0PFAAd5kiSS0tQ5Li483exhYWJk4IiIijeNRoNJkyZh27Zt2rZ69eqxuKB8wyMYJLlL/2sJS9OsCwgLEyNeZ09kQPHx8ejduzd2794NS0tLNGnSBM7OznKHRcUMCwySnKWpESxN+dYiksODBw/g4+ODGzduwNTUFMuXL2dxQbLgtwARURFx+vRpdO7cGc+fP4eTkxN27dqFRo0ayR0WFVMcg0FEVAQEBgaiRYsWeP78OerWrYuQkBAWFyQrFhhEREXAzZs3kZaWhq5du+LUqVNwdXWVOyQq5niKhIioCJg7dy7q1KmDHj16QKnkb0eSH9+FlCtCCCSmqt764yRaRPnp7t27GDhwIFJTUwEARkZG6NWrF4sLKjB4BIP0xkm1iOR15MgRdOvWDTExMShdujTmzp0rd0hEGbDUJb1lN6kWJ9EiMqxly5bB29sbMTExaNiwIcaMGSN3SESZytURDJVKhePHj+PevXvo0aMHbGxs8PTpU9ja2sLa2lrqGKkAe3dSLU6iRWQYaWlpGD16NAICAgAAvXr1wsqVK2Fubi5zZESZ07vAePjwIVq3bo2IiAikpKTg888/h42NDebNm4fk5GQsX77cEHFSAcVJtYgM78WLF+jWrRuOHTsGhUKBOXPmYMKECSzmqUDT+xTJ6NGj4enpiZiYGFhYWGjbO3XqhCNHjkgaHBERAdHR0bh8+TKsra2xe/dufPvttywuqMDT+6fn6dOncebMGZiamuq0u7m54cmTJ5IFRkREb1StWhU7duyAo6MjPvjgA7nDIcoRvY9gaDQaqNUZL0l8/PgxbGxsJAmKiKg4E0Jg4cKFOkeFW7RoweKCChW9C4zPP/8cixYt0j5WKBR4/fo1pk2bhrZt20oZGxFRsZOSkoKBAwfim2++Qbdu3fDff//JHRJRruh9iuTnn39G8+bNUaNGDSQnJ6NHjx64e/cuSpcujU2bNhkiRipA3kywxUm1iAwhKioKnTt3xpkzZ6BUKjF9+nQ4OjrKHRZRruhdYLi4uODq1avYvHkzQkNDodFoMHDgQPTs2VNn0CcVPZxgi8hwrl27Bh8fH0RERMDOzg5bt25Fq1at5A6LKNf0LjBOnjyJxo0bo3///ujfv7+2XaVS4eTJk/jkk08kDZAKjncn2OKkWkTS2LVrF3r16oWEhARUrlwZe/fuRdWqVeUOiyhP9C4wmjdvjsjIyAyH7WJjY9G8efNMB4BS0XPpfy1RysqUl8oRSWD37t1ISEhAy5YtsXXrVtjb28sdElGe6V1gCCEy/VJ58eIFrKysJAmKCj5LU87YSSSVgIAA1KpVC6NGjYKxMSeuo6Ihx+/kzp07A3hz1Ui/fv1gZmamXaZWq3H9+nU0btxY+giJiIqYp0+fYsmSJfj+++9hZGQEc3Nz+Pv7yx0WkaRyXGDY2dkBeHMEw8bGRmdAp6mpKRo1aoRBgwZJHyERURFy6dIldOjQAU+fPoW5uTmmT58ud0hEBpHjAiMwMBAA4O7ujnHjxvF0CBGRnrZs2YJ+/fohOTkZNWrUQO/eveUOichg9J5oa9q0aSwuiikh5I6AqHDSaDSYMmUKvvzySyQnJ6Nt27Y4d+4cKlasKHdoRAaTq9FE27Ztw9atWxEREYHU1FSdZZcvX5YkMCpYhBDotvyc3GEQFToJCQno06cPduzYAQAYN24cfvzxRxgZ8RJvKtr0PoKxZMkS9O/fH46Ojrhy5QoaNGiAUqVK4f79+2jTpo0hYqQCIClNjbDIOABADWdbzn9BlEN3797F/v37YWpqiqCgIMyfP5/FBRULeh/BWLZsGVasWIGvvvoKa9euxYQJE1ChQgVMnToVL1++NESMVMD8MdSLl6gS5VCdOnXw+++/w9nZmVfaUbGi9xGMiIgI7X8SCwsLxMfHAwB69+7Ne5EUE6wtiLK3du1aXLp0Sfu4S5cuLC6o2NG7wChTpgxevHgBAHBzc8P58+cBAOHh4RAcBUhExZharca4cePQr18/dOjQAc+fP5c7JCLZ6F1gtGjRAnv37gUADBw4EGPHjsXnn38OX19fdOrUSfIAiYgKg9jYWPj4+OCnn34C8ObzsVSpUjJHRSQfvQuMFStWYPLkyQCAoUOHIigoCNWrV8eMGTMQEBCgdwDLli2Dh4cHzM3NUa9ePZw6dSrb/ikpKZg8eTLc3NxgZmaGihUrYs2aNXo/LxGRVO7duwcvLy/s378f5ubm2LRpE2bOnAmlUu+PWKIiQ+9BnkqlUuc/Tffu3dG9e3cAwJMnT1C2bNkcb2vLli0YM2YMli1bhiZNmuC3335DmzZtEBYWhvLly2e6Tvfu3fHff/9h9erVqFSpEqKioqBSqfTdDSIiSRw7dgzdu3fHy5cv4eLigt27d8PT01PusIhkJ0l5/ezZM3z99deoVKmSXustXLgQAwcOhJ+fH6pXr45FixbB1dU1yyMhBw8exIkTJ7B//360bNkS7u7uaNCgAQdPGZgQAompvEsuUWZ+/fVXvHz5EvXr10dISAiLC6L/k+MjGK9evcKIESNw6NAhmJiYYOLEiRg5ciSmT5+OBQsWoGbNmnqdqkhNTUVoaCgmTpyo096qVSucPXs203X27NkDT09PzJs3D+vXr4eVlRV8fHwwa9YsnXujvC0lJQUpKSnax3Fxb+Zy0Gg00Gg0OY73fTQaDYQQkm6zIBBCoPtv5xEa8UrbJnXuMlNU8ykn5lRa6flcvXo1qlatiilTpsDCwoL5zQO+R6VliHzqs60cFxiTJk3CyZMn0bdvXxw8eBBjx47FwYMHkZycjAMHDqBZs2Z6BRkdHQ21Wg0nJyeddicnJzx79izTde7fv4/Tp0/D3NwcO3fuRHR0NIYPH46XL19mWdzMmTMHM2bMyND+/PlzJCcn6xVzdjQaDWJjYyGEKFLnXZPS1DrFRW0XK8THvMBrA1+rWlTzKSfmVBqvXr3Cli1b4Ofnh7i4ONjZ2WHMmDGIj4/XXrZPucP3qLQMkU993uM5LjD+/PNPBAYGomXLlhg+fDgqVaqEKlWqYNGiRbmJUevdCZuEEFlO4qTRaKBQKLBhwwbt3V0XLlyIrl27YunSpZkexfjuu+90boMcFxcHV1dXODg4wNbWNk+xZxabg4NDkfqPkZj6/8e3XJzUAqWsTPNlkq2imk85Mad5988//6Bjx464e/curKys0LNnT+ZTQnyPSssQ+TQ3N89x3xwXGE+fPkWNGjUAABUqVIC5uTn8/Pz0j+7/lC5dGkZGRhmOVkRFRWU4qpHO2dkZZcuW1RYXAFC9enUIIfD48WNUrlw5wzpmZmYwMzPL0P7uYFUpKBQKg2xXTm/vi7W5Sb5OcVwU8yk35jT3/vrrL/j6+iI2NhZubm5o2bIl82kAzKm0pM6nPtvJcU+NRgMTExPtYyMjozzdVdXU1BT16tVDcHCwTntwcHCWgzabNGmCp0+f4vXr19q2O3fuQKlUoly5crmOhYgoK0IILFq0CG3btkVsbCw+/vhjXLx4EbVr15Y7NKICLcdHMIQQ6Nevn/ZoQHJyMoYOHZqhyEi/Y2BO+Pv7o3fv3vD09ISXlxdWrFiBiIgIDB06FMCb0xtPnjzBunXrAAA9evTArFmz0L9/f8yYMQPR0dEYP348BgwYkOUgTyKi3EpNTcWIESOwatUqAED//v0REBAAMzMzDkQkeo8cFxh9+/bVedyrV688P7mvry9evHiBmTNnIjIyEh988AH2798PNzc3AEBkZCQiIiK0/a2trREcHIyvv/4anp6eKFWqFLp3747Zs2fnORYionddvnwZgYGBUCqVWLBgAcaMGcMb/RHlUI4LjMDAQIMEMHz4cAwfPjzTZUFBQRnaqlWrluG0ChGRITRq1AgBAQEoV64c2rRpI3c4RIWK3jN5UvEghEBSmpoTbFGxs3fvXlSpUgVVq1YFAAwaNEjmiIgKJxYYlIEQAl2Xn0Powxi5QyHKN0IIzJ07F5MmTULlypVx4cIFlChRQu6wiAotFhiUQVKaOkNx4elmDwuT/LtElSg/JScnw8/PDxs2bAAAtGzZMk9XyRERCwx6j0v/awlLUyNYmBhxcBsVSZGRkejUqRMuXLgAIyMjLFmyJMtxYUSUcywwKFuWpkawNOXbhIqmy5cvo0OHDnj8+DHs7e2xbds2tGjRQu6wiIqEXE3ttX79ejRp0gQuLi54+PAhAGDRokXYvXu3pMERERnSd999h8ePH6NatWq4ePEiiwsiCeldYAQEBMDf3x9t27bFq1evoFa/ucqgRIkSeb4vCRFRflq/fj0GDhyI8+fPo1KlSnKHQ1Sk6F1g/PLLL1i5ciUmT56sc18KT09P3LhxQ9LgiIiklJCQoB3ICQCOjo5YtWqVzv2NiEgaep9cDw8PR926dTO0m5mZISEhQZKgiIik9ujRI3To0AFXrlxBWloa+vXrJ3dIREWa3kcwPDw8cPXq1QztBw4c0N5tlQofIQQSU1X/98fJtahoOX/+POrXr48rV67AwcGBp0OI8oHeRzDGjx+PESNGIDk5GUIIXLx4EZs2bcKcOXO0NwSiwoUTa1FRtn79egwaNAgpKSmoVasW9uzZA3d3d7nDIiry9C4w+vfvD5VKhQkTJiAxMRE9evRA2bJlsXjxYnz55ZeGiJEMLLOJtQBOrkWFm1qtxuTJkzF37lwAQIcOHfD777/D2tpa5siIiodcTXAwaNAgDBo0CNHR0dBoNHB0dJQ6LpJJ+sRaADi5FhVqZ86c0RYXkyZNwqxZs6BU5urKfCLKBb0LjBkzZqBXr16oWLEiSpcubYiYSEacWIuKik8++QQ//PAD3Nzc0KNHD7nDISp29C7nt2/fjipVqqBRo0b49ddf8fz5c0PERUSkt1OnTuHJkyfax9999x2LCyKZ6F1gXL9+HdevX0eLFi2wcOFClC1bFm3btsXGjRuRmJhoiBiJiN5r5cqVaNGiBTp06MDPIqICIFcnJGvWrIkffvgB9+/fx7Fjx+Dh4YExY8agTJkyUsdHRJQtlUqF0aNHY/DgwVCpVKhYsaLcIRERcllgvM3KygoWFhYwNTVFWlqaFDFRPhNC7giIcicmJgZt27bFkiVLAAAzZ87E5s2bYWlpKXNkRJSrAiM8PBzff/89atSoAU9PT1y+fBnTp0/Hs2fPpI6PDEwIgW7Lz8kdBpHe7ty5g0aNGiE4OBiWlpbYtm0bpkyZwiufiAoIvS8X8PLywsWLF1GrVi30799fOw8GFU5JaWqERcYBAGo423LeCyo0Bg8ejDt37sDV1RV79uxBnTp15A6JiN6id4HRvHlzrFq1CjVr1jREPCSjP4Z68dcfFRpr167FqFGjsGLFCjg5OckdDhG9Q+8C44cffjBEHFQAsLaggiw1NRXHjh2Dt7c3AMDNzQ27d++WOSoiykqOCgx/f3/MmjULVlZW8Pf3z7bvwoULJQmMiChddHQ0unbtipMnT2L37t1o37693CER0XvkqMBIv71x+r+JiPLLzZs30b59e4SHh8PGxgZGRhwnRFQY5KjAOHbsWKb/JiIypH379qFHjx6Ij49HhQoVsGfPHo7/Iiok9L5MdcCAAYiPj8/QnpCQgAEDBkgSFBEVb0IIzJ8/Hz4+PoiPj0ezZs1w4cIFFhdEhYjeBcbatWuRlJSUoT0pKQnr1q2TJCgiKt6OHj2KCRMmQAiBwYMH49ChQ7y5IlEhk+OrSOLi4iCEgBAC8fHxMDc31y5Tq9XYv38/b9tORJL47LPPMHr0aFSqVAkjRozg5dNEhVCOC4wSJUpAoVBAoVCgSpUqGZYrFArMmDFD0uCIqPi4du0aypUrh1KlSgEAFi1aJG9ARJQnOS4wjh07BiEEWrRoge3bt6NkyZLaZaampnBzc4OLi4tBgiSiom379u3o06cPGjVqhIMHD8LExETukIgoj3JcYDRr1gzAm/uQlC9fnocsiSjPhBCYPXs2pk6dCgAwNjZGcnIyCwyiIiBHBcb169fxwQcfQKlUIjY2Fjdu3Miyb+3atSULjoiKrsTERPTv3x9bt24FAIwePRoLFiyAsbHeEwwTUQGUo//JderUwbNnz+Do6Ig6depAoVBAZHKPb4VCAbVaLXmQRFS0PHnyBB06dEBoaChMTEywbNky+Pn5yR0WEUkoRwVGeHg4HBwctP8mIsotIQR8fX0RGhqK0qVLY/v27fjkk0/kDouIJJajAsPNzS3TfxMR6UuhUOC3337D4MGD8fvvv8PDw0PukIjIAHI10daff/6pfTxhwgSUKFECjRs3xsOHDyUNjvJOCIHEVFU2fzylRYan0WgQEhKifVyzZk2cPn2axQVREZar27UHBAQAAM6dO4dff/0VixYtwr59+zB27Fjs2LFD8iApd4QQ6Lr8HEIfxsgdChVjr1+/Rq9evbB//34cPnxYezqEV6IRFW16FxiPHj1CpUqVAAC7du1C165dMXjwYDRp0gSffvqp1PFRHiSlqXNcXHi62cPChHepJGk9fPgQPj4+uH79OszMzBAZGSl3SESUT/QuMKytrfHixQuUL18ehw4dwtixYwEA5ubmmd6jhAqGS/9rCUvTrAsICxMj/qIkSZ0+fRqdO3fG8+fP4eTkhF27dqFRo0Zyh0VE+UTvAuPzzz+Hn58f6tatizt37qBdu3YAgJs3b8Ld3V3q+EgilqZGsDTl/AKUP9asWYOhQ4ciLS0NdevWxe7du+Hq6ip3WESUj/Qe5Ll06VJ4eXnh+fPn2L59u/a+AaGhofjqq68kD5CICpfDhw9j4MCBSEtLQ9euXXHq1CkWF0TFkN4/aUuUKIFff/01QztvdEZEwJs7oX711VeoWrUqpkyZAqVS798xRFQE5OqY+atXr7B69WrcunULCoUC1atXx8CBA2FnZyd1fERUCNy7dw9lypSBlZUVFAoFfv/9dxYWRMWc3p8Aly5dQsWKFfHzzz/j5cuXiI6Oxs8//4yKFSvi8uXLhoiRiAqwI0eOoH79+ujbty80Gg0AsLggIv0LjLFjx8LHxwcPHjzAjh07sHPnToSHh+OLL77AmDFjDBAi6UN3Yi1OokWGtXTpUnh7eyMmJgaPHz9GfHy83CERUQGh9ymSS5cuYeXKlTp3PDQ2NsaECRPg6ekpaXCkH06sRfklLS0No0eP1k6617NnT6xatQrm5uYyR0ZEBYXeRzBsbW0RERGRof3Ro0ewsbGRJCjKnawm1uIkWiSlFy9ewNvbGwEBAVAoFJgzZw7Wr1/P4oKIdOh9BMPX1xcDBw7EggUL0LhxYygUCpw+fRrjx4/nZaoFyNsTa3ESLZKKEAI+Pj44e/YsrK2tsWHDBvj4+MgdFhEVQHoXGAsWLIBCoUCfPn2gUqkAACYmJhg2bBh+/PFHyQOk3OHEWmQICoUC8+fPx4ABA/DHH3+gVq1acodERAWU3t9ApqamWLx4MebMmYN79+5BCIFKlSrB0tLSEPERkcyEELh37572HkSNGzfG33//rTMOi4joXTkeg5GYmIgRI0agbNmycHR0hJ+fH5ydnVG7dm0WF0RFVEpKCgYMGIA6derg2rVr2nYWF0T0PjkuMKZNm4agoCC0a9cOX375JYKDgzFs2DBDxkZEMoqKikKLFi0QFBSEpKQkznNDRHrJ8c+QHTt2YPXq1fjyyy8BAL169UKTJk2gVqthZMQrFOQihEBS2pv5LjjvBUnl2rVr8PHxQUREBOzs7LBlyxZ4e3vLHRYRFSI5LjAePXqEpk2bah83aNAAxsbGePr0KW9kJBPOe0GGsHPnTvTu3RsJCQmoXLky9u7di6pVq8odFhEVMjkuMNRqNUxNTXVXNjbWXklC+Y/zXpDUgoOD0blzZwBAy5YtsXXrVtjb28scFREVRjkuMIQQ6NevH8zMzLRtycnJGDp0KKysrLRtO3bskDZCyhHOe0FSaN68OVq1aoWqVati4cKFHMxJRLmW40+Pvn37Zmjr1auXpMFQ7nHeC8qtZ8+eoWTJkjA1NYWxsTH27t2b4WglEZG+cvyNFBgYaMg4iEgGly5dQocOHfDFF19g+fLlUCgULC6ISBK8pzJRMbVlyxY0bdoUT58+xenTpxEXFyd3SERUhMheYCxbtgweHh4wNzdHvXr1cOrUqRytd+bMGRgbG6NOnTqGDZCoiNFoNJgyZQq+/PJLJCcno127djh37hzs7OzkDo2IihBZC4wtW7ZgzJgxmDx5Mq5cuYKmTZuiTZs2md6t9W2xsbHo06cPPvvss3yKlKhoSExMRPfu3TF79mwAwPjx47F7927Y2trKHBkRFTWyFhgLFy7EwIED4efnh+rVq2PRokVwdXVFQEBAtusNGTIEPXr0gJeXVz5FSlT4CSHQq1cv7Ny5E6ampggKCsK8efM4UR4RGYRslx2kpqYiNDQUEydO1Glv1aoVzp49m+V6gYGBuHfvHn7//Xftr7DspKSkICUlRfs4/TyzRqOBRqPJZfQZaTQaCCEk3WZOnvPtf+fncxuaHPks6oQQGDFiBB48eICtW7eicePGzG8e8D0qPeZUWobIpz7bylWBsX79eixfvhzh4eE4d+4c3NzcsGjRInh4eKBDhw452kZ0dDTUajWcnJx02p2cnPDs2bNM17l79y4mTpyIU6dO5fj6/Dlz5mDGjBkZ2p8/f47k5OQcbSMnNBoNYmNjIYSAUpk/B4bSpwgH3uxPUZpcS458FlXR0dEoXbo0NBoNPvroI5w6dQpWVlaIioqSO7RCje9R6TGn0jJEPuPj43PcV+8CIyAgAFOnTsWYMWPw/fffQ61+8yVXokQJLFq0KMcFRrp3J4QSQmQ6SZRarUaPHj0wY8YMVKlSJcfb/+677+Dv7699HBcXB1dXVzg4OEh63lmj0UChUMDBwSHf/mMkpv7/WVQdHByK1DwYcuSzqFGr1fjuu+8QGBiI8+fPw8PDgzmVEN+j0mNOpWWIfJqbm+e4r97fSL/88gtWrlyJjh074scff9S2e3p6Yty4cTneTunSpWFkZJThaEVUVFSGoxrAm6rp0qVLuHLlCkaOHAng/x/+MTY2xqFDh9CiRYsM65mZmenMPppOqVRK/gZWKBQG2W5W3n6e/Hze/JLf+SxK4uLi8NVXX2H//v0A3kwBPnToUOZUYsyn9JhTaUmdT322o3eBER4ejrp162ZoNzMzQ0JCQo63Y2pqinr16iE4OBidOnXStgcHB2d6FMTW1hY3btzQaVu2bBmOHj2Kbdu2wcPDQ4+9ICq67t27Bx8fH4SFhcHc3BxBQUHw9fXleW0iyld6FxgeHh64evUq3NzcdNoPHDiAGjVq6LUtf39/9O7dG56envDy8sKKFSsQERGBoUOHAnhzeuPJkydYt24dlEolPvjgA531HR0dYW5unqGdqLg6fvw4unTpgpcvX8LFxQW7d++Gp6en3GERUTGkd4Exfvx4jBgxAsnJyRBC4OLFi9i0aRPmzJmDVatW6bUtX19fvHjxAjNnzkRkZCQ++OAD7N+/X1u8REZGvndODCJ648iRI2jdujVUKhXq16+PXbt2wcXFRe6wiKiYUgghhL4rrVy5ErNnz8ajR48AAGXLlsX06dMxcOBAyQOUWlxcHOzs7BAbGyv5IM+oqCg4Ojrm6yDPGlP/AgCEzfQucoM88zufhV1ycjKaNWuGihUrYvXq1bCwsNBZzpxKi/mUHnMqLUPkU5/v0Fx9Iw0aNAiDBg1CdHQ0NBoNHB0dcxUoEeVNbGwsbGxsoFQqYW5ujuDgYNjY2GR6JRYRUX7KU0lTunRpFhdEMvnnn3/g6emJqVOnattsbW1ZXBBRgZCrQZ7ZfYDdv38/TwER0fsdPHgQX375JWJjY/H7779jwoQJvJ8IERUoehcYY8aM0XmclpaGK1eu4ODBgxg/frxUcRFRJoQQWLx4Mb755htoNBo0adIEO3bsYHFBRAWO3gXG6NGjM21funQpLl26lOeAiChzqampGD58OFavXg0A6N+/PwICAjKdSI6ISG6SDdNt06YNtm/fLtXmiOgtQgh88cUXWL16NZRKJX766SesXr2axQURFViSFRjbtm1DyZIlpdocEb1FoVCgb9++sLW1xb59++Dv78/BnERUoOl9iqRu3bo6H2xCCDx79gzPnz/HsmXLJA2OqLhLSEiAlZUVAKBnz57w9vZG6dKlZY6KiOj99C4wOnbsqPNYqVTCwcEBn376KapVqyZVXETFmhACc+fORUBAAC5cuIAyZcoAAIsLIio09CowVCoV3N3d4e3trf3AIyJpJScnw8/PDxs2bAAAbNq0CWPHjpU5KiIi/eg1BsPY2BjDhg1DSkqKoeIhKtYiIyPx6aefYsOGDTAyMsLSpUtZXBBRoaT3IM+GDRviypUrhoiFqFgLDQ1F/fr1ceHCBdjb2+PQoUMYPny43GEREeWK3mMwhg8fjm+++QaPHz9GvXr1tAPQ0tWuXVuy4IiKixMnTqBNmzZISkpC9erVsWfPHlSqVEnusIiIci3HBcaAAQOwaNEi+Pr6AgBGjRqlXaZQKCCEgEKhgFqtlj5KoiKuTp06cHd3h7u7OzZt2gQ7Ozu5QyIiypMcFxhr167Fjz/+iPDwcEPGQ1RspKSkwNTUFAqFAnZ2djh69CgcHBxgZGQkd2hERHmW4wJDCAEAcHNzM1gwRMXFo0eP0KFDB/Tp00d7fx9emUVERYlegzw5cyBR3p0/fx7169fHlStXMHfuXMTHx8sdEhGR5PQa5FmlSpX3FhkvX77MU0D0hhACSWnZj2dJTOV4l8Jm/fr1GDRoEFJSUlCrVi3s2bMHNjY2codFRCQ5vQqMGTNmcPBZPhBCoOvycwh9GCN3KCQRjUaDSZMmYe7cuQCADh064Pfff4e1tbXMkRERGYZeBcaXX34JR0dHQ8VC/ycpTa1XceHpZg8LEw4MLKiEEOjatSt27twJAJg0aRJmzZoFpVKyew0SERU4OS4wOP5CHpf+1xKWptkXDxYmRnx9CjCFQoFmzZph//79WLNmDXr06CF3SEREBqf3VSSUvyxNjWBpqvd8aFQAqFQqGBu/ee1GjRqF9u3bo0KFCjJHRUSUP3J8jFaj0fD0CFEOrVq1Ch999BFevXoF4M1RDBYXRFSc8CQwkYRUKhXGjBmDQYMG4caNG1i5cqXcIRERyYLH3okkEhMTA19fXwQHBwMAZs6ciXHjxskcFRGRPFhgEEngzp07aN++Pe7cuQNLS0usW7cOXbp0kTssIiLZsMAoQNIn1+IEWoXLuXPn0LZtW7x69Qqurq7Ys2cP6tSpI3dYRESyYoFRQHByrcKrQoUKsLGxQfXq1bFz5044OTnJHRIRkexYYBQQmU2uxQm0Ci6NRqOdKMvJyQnHjh1DuXLlYGZmJnNkREQFAwuMAih9ci1OoFUwRUdHo2vXrhg4cCB69+4NAKhYsaLMURERFSy8TLUASp9ci8VFwXPz5k00aNAAJ06cwDfffIPXr1/LHRIRUYHEAoMoh/bt2wcvLy+Eh4ejQoUKOHbsGG9WRkSUBRYYRO8hhMD8+fPh4+OD+Ph4NGvWDBcuXEDNmjXlDo2IqMBigUGUDSEE+vfvjwkTJkAIgcGDB+PQoUMoXbq03KERERVoLDCIsqFQKODq6gojIyP88ssvWL58OUxNTeUOi4iowONVJESZEEJoB9nOmDEDnTp1wkcffSRzVEREhQePYBC9Y/v27WjRogWSkpIAAEqlksUFEZGeWGAQ/R8hBGbOnImuXbvi+PHj+PXXX+UOiYio0OIpEiIAiYmJ6N+/P7Zu3QoAGDNmDMaOHStzVEREhRcLDCr2njx5gg4dOiA0NBQmJiZYtmwZ/Pz85A6LiKhQY4FBxdrly5fxxRdfIDIyEqVLl8b27dvxySefyB0WEVGhxwKDirUSJUogNTUVH3zwAfbs2QMPDw+5QyIiKhJYYFCxVqFCBQQHB6NSpUqwsbGROxwioiKDV5FQsfL69Wt06dIFf/75p7atbt26LC6IiCTGIxhUbDx48AA+Pj64ceMGTp06hfDwcFhZWckdFhFRkcQjGFQsnD59GvXr18eNGzfg5OSEPXv2sLggIjIgFhhU5K1evRotWrRAdHQ06tati5CQEDRq1EjusIiIijQWGFRkaTQa+Pv7w8/PD2lpaejatStOnToFV1dXuUMjIiryWGBQkaVUKpGcnAwAmD59OrZs2cLTIkRE+YSDPKlIW7x4Mbp06YLPPvtM7lCIiIoVHsGgIuXo0aPw9fWFSqUCAJiYmLC4ICKSAQsMKjKWLVuGVq1aYevWrVi8eLHc4RARFWs8RSIDIQSS0tQ6bYmp6ix60/ukpaVh9OjRCAgIAAD06tULI0aMkDkqIqLijQVGPhNCoOvycwh9GCN3KEXCy5cv0a1bNxw9ehQKhQJz5szBhAkToFAo5A6NiKhYY4GRz5LS1NkWF55u9rAwMcrHiAqvW7duoX379rh37x6sra2xYcMG+Pj4yB0WERGBBYasLv2vJSxNdYsJCxMj/vrOIZVKhf/++w/u7u7Ys2cPatWqJXdIRET0f1hgyMjS1AiWpnwJcqtWrVrYt28fatSoAQcHB7nDISKit/AqEio0UlJSMHjwYJw5c0bb1qxZMxYXREQFEH8+U6EQFRWFTp064ezZs9i3bx/+/fdfWFpayh0WERFlQfYjGMuWLYOHhwfMzc1Rr149nDp1Ksu+O3bswOeffw4HBwfY2trCy8sLf/31Vz5GS3K4du0a6tevj7Nnz8LOzg6BgYEsLoiICjhZC4wtW7ZgzJgxmDx5Mq5cuYKmTZuiTZs2iIiIyLT/yZMn8fnnn2P//v0IDQ1F8+bN0b59e1y5ciWfI6f8smvXLjRp0gQRERGoVKkSzp8/D29vb7nDIiKi95C1wFi4cCEGDhwIPz8/VK9eHYsWLYKrq6t2wqR3LVq0CBMmTED9+vVRuXJl/PDDD6hcuTL27t2bz5HrTwiBxFQVJ9TKISGE9j4iCQkJ+Oyzz3DhwgVUq1ZN7tCIiCgHZBuDkZqaitDQUEycOFGnvVWrVjh79myOtqHRaBAfH4+SJUtm2SclJQUpKSnax3Fxcdp1NRpNLiLPOhYhRKbbFEKg+2/nERrxKsM6UsZQlKjVaty8eRMAMGLECPz0008wMTFhvvIgu/co6Y/5lB5zKi1D5FOfbclWYERHR0OtVsPJyUmn3cnJCc+ePcvRNn766SckJCSge/fuWfaZM2cOZsyYkaH9+fPn2lt5S0Gj0SA2NhZCCCiVugeGktLUGYqL2i5WiI95gdec8yJTGo0G06dPR5s2bdCpUyfExHDm07zK7j1K+mM+pcecSssQ+YyPj89xX9mvInl3UikhRI4mmtq0aROmT5+O3bt3w9HRMct+3333Hfz9/bWP4+Li4Orqqh0oKhWNRgOFQgEHB4cML2Riqkr774uTWsDS1IgTamXi0qVLWL9+PRYtWqR9HwwaNIgfNBLJ7j1K+mM+pcecSssQ+TQ3N89xX9kKjNKlS8PIyCjD0YqoqKgMRzXetWXLFgwcOBB//PEHWrZsmW1fMzMzmJmZZWhXKpWSv4EVCkWm2337sbW5CSfXysTmzZvRv39/JCcno1q1ahg2bFiW+aTcY06lxXxKjzmVltT51Gc7sr2CpqamqFevHoKDg3Xag4OD0bhx4yzX27RpE/r164eNGzeiXbt2hg6TDEyj0WDKlCn46quvkJycjHbt2qF3795yh0VERHkk609pf39/9O7dG56envDy8sKKFSsQERGBoUOHAnhzeuPJkydYt24dgDfFRZ8+fbB48WI0atRIe/TDwsICdnZ2su0H5c7r16/Rp08f7Ny5EwAwfvx4zJkzB0ZGRhzkRURUyMlaYPj6+uLFixeYOXMmIiMj8cEHH2D//v1wc3MDAERGRurMifHbb79BpVJhxIgRGDFihLa9b9++CAoKyu/wKQ8ePnyIDh064Nq1azA1NcWKFSvQt29fucMiIiKJyD4YYPjw4Rg+fHimy94tGo4fP274gChfPH78GGFhYXB0dMSuXbvg5eUld0hERCQh2QsMKp6aNGmCLVu2oF69eihfvrzc4RARkcQ4TJfyhVqtxuTJk3Hjxg1tW6dOnVhcEBEVUSwwyODi4uLg4+ODH374AR06dEBSUpLcIRERkYHxFAkZ1L179+Dj44OwsDCYm5tjzpw5sLCwkDssIiIyMBYYZDDHjh1D165d8fLlS7i4uGD37t3w9PSUOywiIsoHPEVCBvHbb7+hVatWePnyJerXr4+QkBAWF0RExQgLDJKcRqPBtm3boFKp0KNHD5w4cQIuLi5yh0VERPmIp0hIckqlElu3bsXGjRsxfPhw3tSNiKgY4hEMksQ///yD2bNnax/b29tjxIgRLC6IiIopHsHIB0LIHYFh/fXXX/D19UVsbCxcXFwwYMAAuUMiIiKZ8QiGgQkh0G35ObnDMAghBBYvXoy2bdsiNjYWTZo0wRdffCF3WEREVACwwDCwpDQ1wiLjAAA1nG1hYWIkc0TSSE1NxeDBgzFmzBhoNBr0798fR44cgaOjo9yhERFRAcBTJPnoj6FeRWJMwvPnz9GlSxecOnUKSqUS8+fPx9ixY4vEvhERkTRYYOSjovL9e/nyZZw+fRq2trbYvHkz2rRpI3dIRERUwLDAIL15e3tjxYoVaNKkCapXry53OEREVABxDAa9lxACP//8M8LDw7Vtfn5+LC6IiChLLDAoW8nJyejduzf8/f3h4+OD5ORkuUMiIqJCgKdIJCaEQGKqSvs4MVUtYzR5ExkZiY4dO+LixYswMjLCsGHDYG5uLndYRERUCLDAkJAQAt1/O4/QiFdyh5JnoaGh6NChA548eQJ7e3v88ccf+Oyzz+QOi4iICgmeIpFQskqTZXHh6WZfaObA2Lp1K5o2bYonT56gWrVquHjxIosLIiLSC49gGMil/7WEpen/LygsTIwKxTwRarUaP/30E5KSktC6dWts3rwZdnZ2codFRESFDAsMA7E0NYKlaeFLr5GREXbu3IkVK1ZgypQpMDIqHEddiIioYOEpEsKjR4+wcuVK7WMXFxdMnz6dxQUREeVa4fuJTZI6f/48OnbsiP/++w/29vbo2rWr3CEREVERwCMYxdj69evx6aef4r///kPt2rVRv359uUMiIqIiggVGMaRWqzFx4kT06dMHKSkp6NixI86cOQM3Nze5QyMioiKCBUYxEx8fj06dOmHu3LkAgMmTJ2P79u2wtraWOTIiIipKOAajmAkODsbevXthZmaGNWvWoEePHnKHRO+hVquRlpaW5+1oNBqkpaUhOTkZSiV/W+QV8yk95lRauc2nqampJPlngVHMdO7cGXPmzEGLFi3QoEEDucOhbAgh8OzZM7x69Uqy7Wk0GsTHxxeKOVkKOuZTesyptHKbT6VSCQ8PD5iamubp+VlgFAPr169Hq1at4OTkBACYOHGizBFRTqQXF46OjrC0tMzzB64QAiqVCsbGxvzwlgDzKT3mVFq5yadGo8HTp08RGRmJ8uXL5+l1YIFRhKlUKnzzzTdYsmQJGjdujGPHjuW5IqX8oVartcVFqVKlJNkmP7ylxXxKjzmVVm7z6eDggKdPn0KlUsHExCTXz88Co4iKiYmBr68vgoODAQBt2rTJ0xuF8lf6mAtLS0uZIyGi4ib9h6harWaBQbru3LmD9u3b486dO7C0tMT69evRuXNnucOiXOCvOCLKb1J97rDAKGKCg4PRvXt3vHr1Cq6urtizZw/q1Kkjd1hERFTM8DqgIkStVmPs2LF49eoVvLy8EBISwuKCiiR3d3csWrQo1+sHBQWhRIkSksVTlHz66acYM2ZMvjzXlClTMHjw4Hx5Lvr/6tevjx07dhj8eVhgFCHpd0IdMWIEjh07pr1qhCg/9evXDx07djToc4SEhOT4iymzYsTX1xd37tzJ9fMHBQVBqVRq5wtwcnJC+/btcfPmzVxvs6DYsWMHZs2aZfDn+e+//7B48WJMmjQpw7KzZ8/CyMgIrVu3zrDs+PHjUCgUmV6+XadOHUyfPl2n7cqVK+jWrRucnJxgbm6OKlWqYNCgQXl6/XNi2bJl8PDwgLm5OerVq4dTp069d50NGzbgww8/hKWlJZydndG/f3+8ePFCu3zlypVo2rQp7O3tYW9vj5YtW+LixYs62wgICEDt2rVha2sLOzs7NG3aFAcOHNDpM2XKFEycOBEajUaanc0CC4xCLjo6WqcSrVy5Mn799VeYmZnJGBWRYTk4OORpAKyFhQUcHR3zFIOtrS0iIiLw5MkT/Pnnn0hISEC7du2Qmpqap+2+jxSTrmWnZMmSsLGxMehzAMDq1avh5eUFd3f3DMvWrFmDr7/+GqdPn0ZERESun2Pfvn1o1KgRUlJSsGHDBty6dQvr16+HnZ0dpkyZkofos7dlyxaMGTMGkydPxpUrV9C0aVO0adMm2305ffo0+vTpg4EDB+LmzZv4448/EBISAj8/P22f48eP46uvvsKxY8dw7tw5lC9fHq1atcKTJ0+0fcqVK4cff/wRly5dQkhICD799FN07NhRp/ht164dYmNj8ddffxkmAelEMRMbGysAiNjYWEm3q1arxf2Ix8Lt233C7dt9IiElTdLtZ+bvv/8WHh4eQqlUir/++svgz5ef1Gq1iIyMFGq1Wu5QZJGUlCTCwsJEUlKStk2j0YiElLRc/71OThWvXieK18mpeq+r0WhyHHvfvn1Fhw4dslx+/PhxUb9+fWFqairKlCkjvv32W5GW9v//v8TFxYkePXoIS0tLUaZMGbFw4ULRrFkzMXr0aG0fNzc38fPPP2sfT5s2Tbi6ugpTU1Ph7Owsvv76ayGEEM2aNRMAdP6EECIwMFDY2dnpxLV7925Rr149YWZmJkqVKiU6deqU5T6kr5+amqrNzZ49ewQAcf36dW2/M2fOiKZNmwpzc3NRrlw58fXXX4vXr19rlz99+lS0bdtWmJubC3d3d7Fhw4YM+wZABAQECB8fH2FpaSmmTp2qfb6PPvpImJmZCQ8PDzF9+nSdPGaVEyGEWLp0qahUqZIwMzMTjo6OokuXLtpl7+b65cuXonfv3qJEiRLCwsJCtG7dWty5cydDLg4ePCiqVasmrKyshLe3t3j69GmW+RNCiFq1aolff/1Vp02j0YiYmBhhY2Mj/vnnH+Hr6ytmzJih0+fYsWMCgIiJicmwzQ8//FBMmzZNCCFEQkKCKF26tOjYsWOmz5/Z+lJp0KCBGDp0qE5btWrVxMSJE7NcZ/78+aJChQo6bUuWLBHlypXLch2VSiVsbGzE2rVrM12u0WhEamqqsLe3F6tWrdJZ1q9fP9G7d+9M18vs8yedPt+hHORZSO3btw89evRAfHw8KlSogLJly8odEhlYUpoaNaYa+BdHFsJmesPSNO8fF0+ePEHbtm3Rr18/rFu3Dv/88w8GDRoEc3Nz7aFtf39/nDlzBnv27IGTkxOmTp2Ky5cvZzmeaNu2bfj555+xefNm1KxZE8+ePcO1a9cAvDnc/+GHH2Lw4MEYNGhQlnH9+eef6Ny5MyZPnoz169cjNTUVf/75Z47369WrV9i4cSMAaC/ru3HjBry9vTFr1iysXr0az58/x8iRIzFy5EgEBgYCAPr06YPo6GgcP34cJiYm8Pf3R1RUVIbtT5s2DXPmzMHPP/8MIyMj/PXXX+jVqxeWLFmCpk2b4t69e9pTRtOmTcs2J5cuXcKoUaOwfv16NG7cGC9fvsz28H2/fv1w9+5d7NmzB7a2tvj222/Rtm1bhIWFafc1MTERCxYswPr166FUKtGrVy+MGzcOGzZsyHSbMTEx+Pvvv+Hp6Zlh2R9//IGqVauiatWq6NWrF77++mtMmTJF7ysb/vrrL0RHR2PChAmZLs9uDM7QoUPx+++/Z7v9sLAwlC9fPkN7amoqQkNDM0xo2KpVK5w9ezbL7TVu3BiTJ0/G/v370aZNG0RFRWHbtm1o165dluskJiYiLS0NJUuWzHS5Wq3Gli1bkJCQAC8vL51lDRo0wLx587LbxTxjgVHICCGwYMECfPvttxBCoFmzZti2bRtKly4td2hE77Vs2TK4urri119/hUKhQLVq1fD06VN8++23mDp1KhISErB27Vps3LgRn332GQAgMDAQLi4uWW4zIiICZcqUQcuWLWFiYoLy5ctrp8EvWbIkjIyMYGNjgzJlymS5je+//x5ffvklZsyYoW378MMPs92X2NhY2NvbQwiBxMREAICPjw+qVasGAJg/fz569OihHTBZuXJlLFmyBM2aNUNAQAAePHiAw4cPIyQkRPtFu2rVKlSuXDnDc/Xo0QMDBgzQPu7duzcmTpyIvn37AgAqVKiAWbNmYcKECZg2bVq2OYmIiICVlRW++OIL2NjYwM3NDXXr1s10H9MLizNnzqBx48YA3owTcHV1xa5du9CtWzcAb07bLF++HBUrVgQAjBw5EjNnzswydw8fPoQQItPXNTAwED179gQAtG7dGq9fv8aRI0fQsmXLLLeXVewAtK+HPmbOnIlx48Zl2yer92R0dDTUanWGMXBOTk549uxZlttr3LgxNmzYAF9fXyQnJ0OlUsHHxwe//PJLlutMnDgRZcuWzZCbGzduwMvLC8nJybC2tsaOHTtQo0YNnT5ly5ZFREQENBqNwe77wgKjEElOTsaQIUOwbt06AMDgwYPxyy+/cHbOYsLCxAhhM71zvb7IwyyJFiZGuX7et926dQteXl46z9+kSRO8fv0ajx8/RkxMDNLS0nTuk2NnZ4eqVatmuc1u3bph0aJFqFChAlq3bo22bduiffv2MDbO+cfb1atXsz3CkRkbGxtcuHABAHDy5EnMnz8fy5cv1y4PDQ3Fv//+q/MrXvzfvSHCw8Nx584dGBsb46OPPtIur1SpEuzt7TM817u/9ENDQxESEoLvv/9e26ZWq5GcnIzExMRsc/L555/Dzc1Nu6x169bo1KlTpmNabt26BWNjYzRs2FDbVqpUKVStWhW3bt3StllaWmqLCwBwdnbO9EhMuqSkJACAubm5Tvvt27cREhKiHVdmbGwMX19frFmzRu8CQwihV/+3OTo65nmMzrv/x4QQ2f6/CwsLw6hRozB16lR4e3sjMjIS48ePx9ChQ7F69eoM/efNm4dNmzbh+PHjGfJYtWpVXL16FTExMfjjjz/Qr18/nDhxQqfIsLCwgEajQUpKCiwsLPK0r1lhgVGIbN++HevWrYORkREWLVqEESNGcCKmYkShUOTpNIUQAiolZJ2GObMP2fQvAoVCofPvzPpkxtXVFbdv30ZwcDAOHz6M4cOHY/78+Thx4kSOZyHMzQesUqlEpUqVYGxsjOrVq+PZs2fw9fXFyZMnAby5p8OQIUMwatSoDOuWL18et2/fznS7me2rlZWVzmONRoMZM2ZkOoGeubl5tjmxsbHB5cuXcfz4cRw6dAhTp07F9OnTERISkuG0QVZ5f/d1fDfPb7+WmUk/4hoTEwMHBwdt++rVq6FSqVCuXDmd5zIxMUFMTAzs7e1ha2sL4M0RpHfjffXqFezs7AAAVapUAQD8888/GU4PvE9eTpGULl0aRkZGGY5WREVFZXtl35w5c9CkSROMHz8eAFC7dm1YWVmhadOmmD17NpydnbV9FyxYgB9++AGHDx9G7dq1M2zL1NQUlSpVghACderUweXLl7F48WL89ttv2j4vX76EpaWlwYoLgFeRFCo9evTA2LFjceDAAYwcOZLFBRU6NWrUwNmzZ3W+fM6ePQsbGxuULVsWFStWhImJic6ld3FxcdrD3VmxsLCAj48PlixZguPHj+PcuXO4ceMGgDcftmq1Otv1a9eujSNHjuRhz4CxY8fi2rVr2LlzJwDgo48+ws2bN1GpUqUMf6ampqhWrRpUKhWuXLmi3ca///6bo7vnfvTRR7h9+3am204/3J1dToyNjdGyZUvMmzcP169fx4MHD3D06NEMz1OjRg2oVCrtkRoAePHiBe7cuYPq1avnOlcVK1aEra0twsLCtG0qlQrr16/HvHnzcOXKFVy9ehVXr17FtWvX4Obmpj0SVLlyZSiVSoSEhOhsMzIyEk+ePNEe7WrVqhVKly6d5TiD7PI8c+ZM7fNn9ZfVKRJTU1PUq1dPe5uGdMHBwdrTTJlJTEzMcKrCyOjNkcO3/7/Mnz8fs2bNwsGDBzMdw5IZIQRSUlJ02v7++2+do2eGwCMYBdz+/fvx8ccfw9bWFgqFAgsXLpQ7JKL3io2NxdWrV3XaSpYsieHDh2PRokX4+uuvMXLkSNy+fRvTpk2Dv78/lEolbGxs0LdvX4wfPx4lS5aEo6Mjpk2bBqVSmWVBHRQUBLVajYYNG2qnxrewsICbmxuAN/NgnDx5El9++SXMzMwyHa80bdo0fPbZZ6hYsSK+/PJLqFQqHDhwIMsBgpmxtbWFn58fpk2bho4dO+Lbb79Fo0aNMGLECAwaNAhWVla4desWgoOD8csvv6BatWpo2bIlBg8ejICAAJiYmOCbb76BhYXFe388TJ06FV988QVcXV3RrVs3KJVKXL9+HTdu3MDs2bOzzcm+fftw//59fPLJJ7C3t8f+/fuh0WgyPQ1VuXJldOjQAYMGDcJvv/0GGxsb7Xn/Dh065Dg371IqlWjZsiVOnz6tnTNl3759iImJQf/+/VGqVCmdHHTt2hWrV6/GyJEjYWNjgyFDhuCbb76BsbExPvzwQzx9+hSTJ09G9erV0apVKwBvjvqsWrUK3bp1g4+PD0aNGoVKlSohOjoaW7duRUREBDZv3pxpfHk9ReLv74/evXvD09MTXl5eWLFiBSIiIjB06FBtn++++w5PnjzRnvJu3749Bg0ahICAAO0pkjFjxqBBgwbaYmbevHmYMmUKNm7cCHd3d+1REmtra1hbWwMAJk2ahDZt2sDV1RVxcXHYuHEjjh8/joMHD+rEeOrUKW2uDOa915kUMYXlMlWNRiNmzpwpAIi2bdsKlUolUaSFAy9TzfoysdxKv2RNn0tOc6Nv374ZLg0FIPr27SuEyN1lqg0aNNC5xO/tSzl37twpGjZsKGxtbYWVlZVo1KiROHz4sLbvuXPnRO3atYWZmVm2l6lu375d1KlTR5iamorSpUuLzp07Z7mPmV2mKoQQDx8+FMbGxmLLli1CCCEuXrwoPv/8c2FtbS2srKxE7dq1xffff6/t//TpU9GmTRthZmYm3NzcxMaNG4Wjo6NYvny5tg8AsXPnzgwxHDx4UDRu3FhYWFgIW1tb0aBBA7FixYr35uTUqVOiWbNmwt7eXlhYWIjatWtr4xUi68tU7ezshIWFhfD29s70MtW37dy5U7zv6+XgwYOibNmy2v/jX3zxhWjbtm2m79HQ0FABQISGhgohhEhOThYzZ84U1atXFxYWFsLNzU3069dPREZGZniekJAQ0blzZ+Hg4CDMzMxEpUqVxODBg8Xdu3ezjS+vli5dKtzc3ISpqan46KOPxIkTJ3SW9+3bVzRr1kynbcmSJaJGjRrCwsJCODs7i549e4rHjx9rl7u5uWX6fyv90lwhhBgwYID2eR0cHESLFi0yTGPw+PFjYWJiIh49epRp7FJdpsoCQyJSFhgJCQmie/fu2jfP6NGjdT6AiwMWGIW3wJDa69evhZ2dXYbr+OVmiHw+evRIANApkIoqjUYjGjRoIDZu3KjTVhjfowVVVvkcN26cGDRoUJbrcR6MIurJkyfo0KEDQkNDYWxsjGXLluk9up2oMLty5Qr++ecfNGjQALGxsdrLHfNySL6gOnr0KF6/fo1atWohMjISEyZMgLu7Oz755BO5QzM4hUKBFStW4Pr163KHUuw4Ojq+9zJcKbDAKEAuXLiAjh074tmzZyhVqhS2b9+OZs2ayR0WUb5bsGABbt++rR0wd+rUqSI510taWhomTZqE+/fvw8bGRjsXQk6vfinsPvzww/fON0LSS79SxdBYYBQQKpUKvXr1wrNnz/DBBx9gz5498PDwkDssonxXt25dhIaGyh1GvvD29oa3d+7nNiEqyHiZagFhbGyMLVu2wNfXF2fPnmVxQUREhRoLDBnFx8fj+PHj2scfffQRNm/enC93MiQiIjIkFhgyefDgAZo0aYLWrVvrTCpERERUFLDAkMHp06dRv3593LhxAyVKlMjTnPlEREQFEQuMfLZmzRq0aNEC0dHRqFu3LkJCQnRuJERERFQUsMDIJ2q1Gt988w0GDhyItLQ0dO3aFadOnYKrq6vcoREREUmOBUY+WbdunfY+ItOnT8eWLVsy3CGRiPLO3d0dixYtkjuMLOVXfA8ePIBCodC5J8yZM2dQq1YtmJiYoGPHjjh+/DgUCkWObrBGpC8WGPmkb9+++Oqrr7B161btzZuIiqJ+/fpBoVBAoVDA2NgY5cuXx7BhwxATEyN3aAYXFxeHyZMno1q1ajA3N0eZMmXQsmVL7NixI9/HWrm6uiIyMhIffPCBts3f3x916tRBeHg4goKC0LhxY0RGRmpvcU4kJU60ZUDnzp1D3bp1YW5uDqVSiY0bN8odElG+aN26NQIDA6FSqRAWFoYBAwbg1atX2LRpk9yhGcyrV6/w8ccfIzY2FrNnz0b9+vVhbGyMEydOYMKECWjRogVKlCiRb/EYGRmhTJkyOm337t3D0KFDUa5cOW3bu330lZqaClNT0zxtg4om/ow2kBXLA9C0aVMMGjSIV4mQpBISErL8S05OznHfpKSkHPXNDTMzM5QpUwblypVDq1at4Ovri0OHDmmXq9VqDBw4EB4eHrCwsEDVqlWxePFinW3069cPHTt2xIIFC+Ds7IxSpUphxIgRSEtL0/aJiopC+/btYWFhAQ8PD2zYsCFDLBEREejQoQOsra1ha2uL7t2747///tMunz59OurUqYM1a9agfPnysLa2xrBhw6BWqzFv3jyUKVMGjo6O+P7777Pd50mTJuHBgwe4cOEC+vbtixo1aqBKlSoYNGgQrl69qr2d9rsWLlyIWrVqwcrKCq6urhg+fDhev36tXf7w4UO0b98e9vb2sLKyQs2aNbF//34AQExMDHr27AkHBwdYWFigcuXKCAwMBKB7iiT93y9evMCAAQOgUCgQFBSU6SmSs2fP4pNPPoGFhQVcXV0xatQonfeBu7s7Zs+ejX79+sHOzo73SqIsyV5gLFu2DB4eHjA3N9fecyA7J06cQL169WBubo4KFSpg+fLl+RRpzgi1Ci8OLcPY0aOgVqsBvJkGnEgq1tbWWf516dJFp6+jo6N2mY2NDezt7WFjYwNra2u0adNGp6+7u3um28yr+/fv4+DBgzr319BoNChXrhy2bt2KsLAwTJ06FZMmTcLWrVt11j127Bju3buHY8eOYe3atQgKCkJQUJB2eb9+/fDgwQMcPXoU27Ztw7JlyxAVFaVdLoRAx44d8fLlS5w4cQLBwcG4d+8efH19dZ7n3r17OHDgAA4ePIhNmzZhzZo1aNeuHR4/fowTJ05g7ty5+N///ofz589nuo8ajQabN29Gz5494eLikmG5tbU1jI0zP2CsVCqxZMkS/P3331i7di2OHj2KCRMmaJePGDECKSkpOHnyJG7cuIG5c+dqX5cpU6YgLCwMBw4cwK1btxAQEJDpPVvST5fY2tpi0aJFiIyMzJADALhx4wa8vb3RuXNnXL9+HVu2bMHp06cxcuRInX7z58/HBx98gNDQUEyZMiXT/SKS9XbtmzdvFiYmJmLlypUiLCxMjB49WlhZWYmHDx9m2v/+/fvC0tJSjB49WoSFhYmVK1cKExMTsW3bthw/pyFv1x567YYwK19bABAKhUL8+OOPvO1wLvF27VnfLhlAln9t27bV6WtpaZll32bNmun0LV26dKb99NW3b19hZGQkrKyshLm5uXY7CxcuzHa94cOHiy5duuhsx83NTahUKm1bt27dhK+vrxBCiNu3bwsA4vz589rlt27dEgDEzz//LIQQ4tChQ8LIyEhERERo+9y8eVMAEBcvXhRCCDFt2jRhaWkp4uLitH28vb2Fu7u7zvuvatWqYs6cOdrHb98K+7///svRPgohhJubmza+zGzdulWUKlVK+7hWrVpi+vTpmfZt37696N+/f6bLwsPDBQBx5coVbZudnZ0IDAzUPj527JgAIGJiYoQQQvTu3VsMHjxYZzunTp0SSqVS+150c3MTHTt2zGYPc4+3a5dWbvNZJG7XvnDhQgwcOBB+fn4AgEWLFuGvv/5CQEAA5syZk6H/8uXLUb58ee0I7OrVq+PSpUtYsGBBhl9u+UkIgcvXb6Bzhw5IiXgIhakFtm7aiK6dO8oWExVdbx8+f5eRkZHO43d/zatUKhgbG0OhUGQYaPzgwQPJYmzevDkCAgKQmJiIVatW4c6dO/j66691+ixfvhyrVq3Cw4cPkZSUhNTUVNSpU0enT82aNXX2ydnZGTdu3AAA3Lp1C8bGxvD09NQur1atms44h1u3bsHV1VXncvAaNWqgRIkSuHXrFurXrw/gzdGbt6fod3JygpGRkU6OnJycdPL5NvF/p0EVCkVO0qPj2LFj+OGHHxAWFoa4uDioVCokJycjISEBVlZWGDVqFIYNG4ZDhw6hZcuW6NKlC2rXrg0AGDZsGLp06YLLly+jVatW6NixIxo3bqx3DOlCQ0Px77//6pxqEkJAo9EgPDwc1atXBwCdnBNlRbYCIzU1FaGhoZg4caJOe6tWrXD27NlM1zl37hxatWql0+bt7Y3Vq1cjLS0t01scp6SkICUlRfs4Li4OwJtDmhqNJq+7AQCITUiC16feUMX+ByM7Jzh2mYLWbdtKtv3iSKPRaD/YiqP0/U//e5ulpWW2677d/92+7/4/ya5vZn1yysrKChUrVgQALF68GC1atMD06dMxa9YsAMDWrVsxduxYLFiwAF5eXrCxscH8+fNx8eJFneczMTHJ8PyZvTfe7ZOeN41GA4VCkenyt/u9+zwKhSLTNrVardOW/u/SpUvD3t4eYWFhOcpX+vM+fPgQbdu2xZAhQzBz5kyULFkSp0+fhp+fH1JTU2FpaYmBAweiVatW+PPPPxEcHIw5c+ZgwYIF+Prrr9G6dWs8ePAAf/75J44cOYLPPvsMw4cPx4IFCzLs47vPnVkeNBoNBg8ejFGjRmWIuXz58tr+lpaWBhtb9m5slDe5yefb74d3P4P1+UyWrcCIjo6GWq2Gk5OTTruTkxOePXuW6TrPnj3LtL9KpUJ0dDScnZ0zrDNnzhzMmDEjQ/vz588zDIjLraQ0NUq1GY3Yc1tQ2mcC6lZyQXzMC7zOxa8ZekOj0SA2NhZCiGJ5SW9aWho0Gg1UKpVkY3iEENpxQbn5pZ1T6R9Kb8c9efJktG/fHoMGDYKLiwtOnjwJLy8vDB48WNvn3r172qMsWW0n/YNPpVKhcuXKUKlUuHDhgvZIxO3bt/Hq1SvtelWrVkVERATCw8O1RzHCwsIQGxuLKlWqQKVSaQuWt5/nfc+d/vjtfHbr1g0bNmzA5MmTM4zDSEhIgJmZmXYcRvq2L1y4AJVKhblz52rf55s3bwYAndfe2dkZfn5+8PPzw+TJk7Fy5UoMGzYMAGBvb49evXqhV69eaNy4MSZOnIgff/xRu+6776G39+vtcWIqlQp16tTBzZs34e7unulr+/ZrY4ixZfn1Hi0ucpvP9P8XL168yPDDPT4+Psfbkf0y1Xd3WgiRbSIy659Ze7rvvvsO/v7+2sdxcXFwdXWFg4MDbG1tcxt2hhjurB6L5897wcHBAVZmJvzPkUfpvzwdHByKZYGRnJyM+Ph4GBsbZzk4MLcyO9InJaVSCaVSqRP3Z599hpo1a2LevHn49ddfUblyZfz+++84cuQIPDw8sH79ely6dAkeHh7a9TLbztvza9SsWROtW7fGsGHD8Ntvv8HY2Bhjx46FhYWFdj1vb2/Url0b/fr1w88//wyVSoURI0agWbNm2in6lUqldpvZ7cPbz/229HzOmTMHJ0+exMcff4zZs2fD09MTJiYmOHXqFH788UdcvHhRe/omfdvpRU5AQADat2+PM2fOYOXKlQCgfe3HjBmDNm3aoEqVKoiJicGJEydQo0YNGBsbY+rUqahXrx5q1qyJlJQUHDhwANWrV9d537z7Hnp7v9JPP6X3mThxIry8vDB69GgMGjQIVlZWuHXrFoKDg/HLL79kug1DMPR7tLjRN5/GxsZQKpUoVaoUzM3NdZa9+zjb7ej1rBIqXbo0jIyMMhytiIqKynCUIl2ZMmUy7W9sbIxSpUpluo6ZmRnMzMwytKd/gEjF2lyBRDMTWJubFssvRENIHydQHPOZ/qWX/ieFt4v3/CiA330Of39/9O/fHxMnTsSwYcNw7do1fPnll1AoFPjqq68wfPhwHDhwIMN6bz9+N/7AwED4+fnh008/hZOTE2bPno0pU6bo5G7Xrl34+uuv0axZMyiVSrRu3Rq//PJLhm1llpPMYklvezefJUuWxPnz5/Hjjz/i+++/x8OHD2Fvb49atWph/vz5KFGihE5/hUKBunXrYuHChZg3bx4mTZqETz75BHPmzEGfPn20fTQaDUaOHInHjx/D1tYWrVu3xs8//wyFQgEzMzPt5bEWFhZo2rQpNm/erBPnu++hd5e93fbhhx/ixIkTmDx5Mj755BMIIVCxYkX4+vpmuQ0p5fd7tKjLbT7TX9/MPn/1+TxWCBlPdDVs2BD16tXDsmXLtG01atRAhw4dMh3k+e2332Lv3r0ICwvTtg0bNgxXr17FuXPncvSccXFxsLOzQ2xsrGRHMIA3v7ijoqLg6OhYLL8QpVbc85mcnIzw8HDtJdxSeHeQJ+UN8yk95lRauc1ndp8/+nyHyvrJ7e/vj1WrVmHNmjW4desWxo4di4iICAwdOhTAm9Mbffr00fYfOnQoHj58CH9/f9y6dQtr1qzB6tWrMW7cOLl2gYiIiDIh6xgMX19fvHjxAjNnztTOmb9//364ubkBACIjIxEREaHt7+Hhgf3792Ps2LFYunQpXFxcsGTJElkvUSUiIqKMZD1FIgeeIikcins+eYqk4GM+pcecSqtYnyIhIiKiookFBlEBVswOMBJRASDV5w4LDKICKP269cTERJkjIaLiJjU1FUDGWw/oS/aJtogoIyMjI5QoUUJ77wtLS8s8n5Pm+W1pMZ/SY06llZt8ajQaPH/+HJaWlnmeTI0FBlEBVaZMGQDI8gZb+kq/t0D6JF6UN8yn9JhTaeU2n0qlEuXLl8/za8ACg6iAUigUcHZ2hqOjI9LS0vK8vfR7C5QqVapYXpkjNeZTesyptHKbT1NTaWakZoFBVMAZGRnl+Vwo8ObDxsTEBObm5vzwlgDzKT3mVFpy55OvIBEREUmOBQYRERFJjgUGERERSa7YjcFIn0AkLi5O0u1qNBrEx8fz3KFEmE/pMafSYj6lx5xKyxD5TP/uzMlkXMWuwIiPjwcAuLq6yhwJERFR4RQfHw87O7ts+xS7m51pNBo8ffoUNjY2kl5nHRcXB1dXVzx69EjSm6gVV8yn9JhTaTGf0mNOpWWIfAohEB8fDxcXl/ceFSl2RzCUSiXKlStnsO3b2tryP4aEmE/pMafSYj6lx5xKS+p8vu/IRTqe5CIiIiLJscAgIiIiybHAkIiZmRmmTZsGMzMzuUMpEphP6TGn0mI+pcecSkvufBa7QZ5ERERkeDyCQURERJJjgUFERESSY4FBREREkmOBQURERJJjgZFDy5Ytg4eHB8zNzVGvXj2cOnUq2/4nTpxAvXr1YG5ujgoVKmD58uX5FGnhoU9Od+zYgc8//xwODg6wtbWFl5cX/vrrr3yMtuDT9z2a7syZMzA2NkadOnUMG2AhpG9OU1JSMHnyZLi5ucHMzAwVK1bEmjVr8inawkHfnG7YsAEffvghLC0t4ezsjP79++PFixf5FG3BdvLkSbRv3x4uLi5QKBTYtWvXe9fJ1+8mQe+1efNmYWJiIlauXCnCwsLE6NGjhZWVlXj48GGm/e/fvy8sLS3F6NGjRVhYmFi5cqUwMTER27Zty+fICy59czp69Ggxd+5ccfHiRXHnzh3x3XffCRMTE3H58uV8jrxg0jef6V69eiUqVKggWrVqJT788MP8CbaQyE1OfXx8RMOGDUVwcLAIDw8XFy5cEGfOnMnHqAs2fXN66tQpoVQqxeLFi8X9+/fFqVOnRM2aNUXHjh3zOfKCaf/+/WLy5Mli+/btAoDYuXNntv3z+7uJBUYONGjQQAwdOlSnrVq1amLixImZ9p8wYYKoVq2aTtuQIUNEo0aNDBZjYaNvTjNTo0YNMWPGDKlDK5Rym09fX1/xv//9T0ybNo0Fxjv0zemBAweEnZ2dePHiRX6EVyjpm9P58+eLChUq6LQtWbJElCtXzmAxFlY5KTDy+7uJp0jeIzU1FaGhoWjVqpVOe6tWrXD27NlM1zl37lyG/t7e3rh06RLS0tIMFmthkZucviv9NsQlS5Y0RIiFSm7zGRgYiHv37mHatGmGDrHQyU1O9+zZA09PT8ybNw9ly5ZFlSpVMG7cOCQlJeVHyAVebnLauHFjPH78GPv374cQAv/99x+2bduGdu3a5UfIRU5+fzcVu5ud6Ss6OhpqtRpOTk467U5OTnj27Fmm6zx79izT/iqVCtHR0XB2djZYvIVBbnL6rp9++gkJCQno3r27IUIsVHKTz7t372LixIk4deoUjI35MfCu3OT0/v37OH36NMzNzbFz505ER0dj+PDhePnyJcdhIHc5bdy4MTZs2ABfX18kJydDpVLBx8cHv/zyS36EXOTk93cTj2Dk0Lu3dhdCZHu798z6Z9ZenOmb03SbNm3C9OnTsWXLFjg6OhoqvEInp/lUq9Xo0aMHZsyYgSpVquRXeIWSPu9RjUYDhUKBDRs2oEGDBmjbti0WLlyIoKAgHsV4iz45DQsLw6hRozB16lSEhobi4MGDCA8Px9ChQ/Mj1CIpP7+b+NPlPUqXLg0jI6MMFXZUVFSGSjBdmTJlMu1vbGyMUqVKGSzWwiI3OU23ZcsWDBw4EH/88QdatmxpyDALDX3zGR8fj0uXLuHKlSsYOXIkgDdfjkIIGBsb49ChQ2jRokW+xF5Q5eY96uzsjLJly+rcyrp69eoQQuDx48eoXLmyQWMu6HKT0zlz5qBJkyYYP348AKB27dqwsrJC06ZNMXv27GJ/NFhf+f3dxCMY72Fqaop69eohODhYpz04OBiNGzfOdB0vL68M/Q8dOgRPT0+YmJgYLNbCIjc5Bd4cuejXrx82btzIc7Bv0Teftra2uHHjBq5evar9Gzp0KKpWrYqrV6+iYcOG+RV6gZWb92iTJk3w9OlTvH79Wtt2584dKJVKlCtXzqDxFga5yWliYiKUSt2vKSMjIwD//5c35Vy+fzcZZOhoEZN+adXq1atFWFiYGDNmjLCyshIPHjwQQggxceJE0bt3b23/9EuBxo4dK8LCwsTq1at5meo79M3pxo0bhbGxsVi6dKmIjIzU/r169UquXShQ9M3nu3gVSUb65jQ+Pl6UK1dOdO3aVdy8eVOcOHFCVK5cWfj5+cm1CwWOvjkNDAwUxsbGYtmyZeLevXvi9OnTwtPTUzRo0ECuXShQ4uPjxZUrV8SVK1cEALFw4UJx5coV7WW/cn83scDIoaVLlwo3NzdhamoqPvroI3HixAntsr59+4pmzZrp9D9+/LioW7euMDU1Fe7u7iIgICCfIy749Mlps2bNBIAMf3379s3/wAsofd+jb2OBkTl9c3rr1i3RsmVLYWFhIcqVKyf8/f1FYmJiPkddsOmb0yVLlogaNWoICwsL4ezsLHr27CkeP36cz1EXTMeOHcv2c1Hu7yberp2IiIgkxzEYREREJDkWGERERCQ5FhhEREQkORYYREREJDkWGERERCQ5FhhEREQkORYYREREJDkWGERERCQ5FhhERUxQUBBKlCghdxi55u7ujkWLFmXbZ/r06ahTp06+xENEucMCg6gA6tevHxQKRYa/f//9V+7QEBQUpBOTs7MzunfvjvDwcEm2HxISgsGDB2sfKxQK7Nq1S6fPuHHjcOTIEUmeLyvv7qeTkxPat2+Pmzdv6r2dwlzwEeUWCwyiAqp169aIjIzU+fPw8JA7LABv7sgaGRmJp0+fYuPGjbh69Sp8fHygVqvzvG0HBwdYWlpm28fa2togt5d+19v7+eeffyIhIQHt2rVDamqqwZ+bqLBjgUFUQJmZmaFMmTI6f0ZGRli4cCFq1aoFKysruLq6Yvjw4Tq3CH/XtWvX0Lx5c9jY2MDW1hb16tXDpUuXtMvPnj2LTz75BBYWFnB1dcWoUaOQkJCQbWwKhQJlypSBs7MzmjdvjmnTpuHvv//WHmEJCAhAxYoVYWpqiqpVq2L9+vU660+fPh3ly5eHmZkZXFxcMGrUKO2yt0+RuLu7AwA6deoEhUKhffz2KZK//voL5ubmePXqlc5zjBo1Cs2aNZNsPz09PTF27Fg8fPgQt2/f1vbJ7vU4fvw4+vfvj9jYWO2RkOnTpwMAUlNTMWHCBJQtWxZWVlZo2LAhjh8/nm08RIUJCwyiQkapVGLJkiX4+++/sXbtWhw9ehQTJkzIsn/Pnj1Rrlw5hISEIDQ0FBMnToSJiQkA4MaNG/D29kbnzp1x/fp1bNmyBadPn8bIkSP1isnCwgIAkJaWhp07d2L06NH45ptv8Pfff2PIkCHo378/jh07BgDYtm0bfv75Z/z222+4e/cudu3ahVq1amW63ZCQEABAYGAgIiMjtY/f1rJlS5QoUQLbt2/XtqnVamzduhU9e/aUbD9fvXqFjRs3AoA2f0D2r0fjxo2xaNEi7ZGQyMhIjBs3DgDQv39/nDlzBps3b8b169fRrVs3tG7dGnfv3s1xTEQFmsHu00pEuda3b19hZGQkrKystH9du3bNtO/WrVtFqVKltI8DAwOFnZ2d9rGNjY0ICgrKdN3evXuLwYMH67SdOnVKKJVKkZSUlOk6727/0aNHolGjRqJcuXIiJSVFNG7cWAwaNEhnnW7duom2bdsKIYT46aefRJUqVURqamqm23dzcxM///yz9jEAsXPnTp0+795eftSoUaJFixbax3/99ZcwNTUVL1++zNN+AhBWVlbC0tJSeytsHx+fTPune9/rIYQQ//77r1AoFOLJkyc67Z999pn47rvvst0+UWFhLG95Q0RZad68OQICArSPraysAADHjh3DDz/8gLCwMMTFxUGlUiE5ORkJCQnaPm/z9/eHn58f1q9fj5YtW6Jbt26oWLEiACA0NBT//vsvNmzYoO0vhIBGo0F4eDiqV6+eaWyxsbGwtraGEAKJiYn46KOPsGPHDpiamuLWrVs6gzQBoEmTJli8eDEAoFu3bli0aBEqVKiA1q1bo23btmjfvj2MjXP/cdSzZ094eXnh6dOncHFxwYYNG9C2bVvY29vnaT9tbGxw+fJlqFQqnDhxAvPnz8fy5ct1+uj7egDA5cuXIYRAlSpVdNpTUlLyZWwJUX5ggUFUQFlZWaFSpUo6bQ8fPkTbtm0xdOhQzJo1CyVLlsTp06cxcOBApKWlZbqd6dOno0ePHvjzzz9x4MABTJs2DZs3b0anTp2g0WgwZMgQnTEQ6cqXL59lbOlfvEqlEk5OThm+SBUKhc5jIYS2zdXVFbdv30ZwcDAOHz6M4cOHY/78+Thx4oTOqQd9NGjQABUrVsTmzZsxbNgw7Ny5E4GBgdrlud1PpVKpfQ2qVauGZ8+ewdfXFydPngSQu9cjPR4jIyOEhobCyMhIZ5m1tbVe+05UULHAICpELl26BJVKhZ9++glK5ZshVFu3bn3velWqVEGVKlUwduxYfPXVVwgMDESnTp3w0Ucf4ebNmxkKmfd5+4v3XdWrV8fp06fRp08fbdvZs2d1jhJYWFjAx8cHPj4+GDFiBKpVq4YbN27go48+yrA9ExOTHF2d0qNHD2zYsAHlypWDUqlEu3bttMtyu5/vGjt2LBYuXIidO3eiU6dOOXo9TE1NM8Rft25dqNVqREVFoWnTpnmKiaig4iBPokKkYsWKUKlU+OWXX3D//n2sX78+wyH7tyUlJWHkyJE4fvw4Hj58iDNnziAkJET7Zf/tt9/i3LlzGDFiBK5evYq7d+9iz549+Prrr3Md4/jx4xEUFITly5fj7t27WLhwIXbs2KEd3BgUFITVq1fj77//1u6DhYUF3NzcMt2eu7s7jhw5gmfPniEmJibL5+3ZsycuX76M77//Hl27doW5ubl2mVT7aWtrCz8/P0ybNg1CiBy9Hu7u7nj9+jWOHDmC6OhoJCYmokqVKujZsyf69OmDHTt2IDw8HCEhIZg7dy7279+vV0xEBZacA0CIKHN9+/YVHTp0yHTZwoULhbOzs7CwsBDe3t5i3bp1AoCIiYkRQugOKkxJSRFffvmlcHV1FaampsLFxUWMHDlSZ2DjxYsXxeeffy6sra2FlZWVqF27tvj++++zjC2zQYvvWrZsmahQoYIwMTERVapUEevWrdMu27lzp2jYsKGwtbUVVlZWolGjRuLw4cPa5e8O8tyzZ4+oVKmSMDY2Fm5ubkKIjIM809WvX18AEEePHs2wTKr9fPjwoTA2NhZbtmwRQrz/9RBCiKFDh4pSpUoJAGLatGlCCCFSU1PF1KlThbu7uzAxMRFlypQRnTp1EtevX88yJqLCRCGEEPKWOERERFTU8BQJERERSY4FBhEREUmOBQYRERFJjgUGERERSY4FBhEREUmOBQYRERFJjgUGERERSY4FBhEREUmOBQYRERFJjgUGERERSY4FBhEREUnu/wFWwpeEWxFHswAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 600x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr, tpr, thresholds=roc_curve(y_val, y_proba_xg)\n",
    "roc_auc=auc(fpr, tpr)\n",
    "\n",
    "plt.figure(figsize=(6,5))\n",
    "plt.plot(fpr, tpr, label=f'Logistic Regression (AUC = {roc_auc:.3f})')\n",
    "plt.plot([0, 1], [0, 1], 'k--', label='Random Classifier')\n",
    "\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curve')\n",
    "plt.legend(loc='lower right')\n",
    "plt.grid(alpha=0.3)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "id": "b3b0a707-e1f0-49a2-89f8-f41956605ea6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['xgb_model.pkl']"
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(xg,'xgb_model.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "id": "9e6672b8-950d-44e6-bc91-f40827901523",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['feature_names.pkl']"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(x_train.columns.tolist(), \"feature_names.pkl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "id": "eba29405-6ea3-4062-a592-6c8dd1235250",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['threshold.pkl']"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(0.35, \"threshold.pkl\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "943c9955-b7eb-42d0-ad57-05f5d765c616",
   "metadata": {},
   "source": [
    "### 5th CatBoostClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "id": "614b17be-090c-44e9-a736-106134bead4b",
   "metadata": {},
   "outputs": [],
   "source": [
    "cb=CatBoostClassifier(iterations=2000,learning_rate=0.03,depth=6,l2_leaf_reg=6,loss_function=\"Logloss\",eval_metric=\"AUC\",class_weights=[1.0, 6.0],random_seed=42,verbose=200)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "id": "48b65e48-4bc0-4ebd-885f-d84d3a87d36b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\ttotal: 147ms\tremaining: 4m 53s\n",
      "200:\ttotal: 414ms\tremaining: 3.7s\n",
      "400:\ttotal: 641ms\tremaining: 2.56s\n",
      "600:\ttotal: 863ms\tremaining: 2.01s\n",
      "800:\ttotal: 1.08s\tremaining: 1.62s\n",
      "1000:\ttotal: 1.3s\tremaining: 1.3s\n",
      "1200:\ttotal: 1.53s\tremaining: 1.02s\n",
      "1400:\ttotal: 1.76s\tremaining: 752ms\n",
      "1600:\ttotal: 1.98s\tremaining: 494ms\n",
      "1800:\ttotal: 2.21s\tremaining: 244ms\n",
      "1999:\ttotal: 2.44s\tremaining: 0us\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<catboost.core.CatBoostClassifier at 0x1c53a99c590>"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cb.fit(x_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "id": "3b7e6669-1e1e-45bd-922d-ca6d319778cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "cbtahmin=cb.predict(x_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "id": "15ea83ba-ae35-4837-bc28-c896e552b900",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_proba_cb=cb.predict_proba(x_val)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "id": "19267ecd-79f4-4bab-8848-6886ae47c581",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC: 0.8155\n"
     ]
    }
   ],
   "source": [
    "roc_auc=roc_auc_score(y_val, y_proba_cb)\n",
    "print(f\"ROC AUC: {roc_auc:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "id": "a980f191-6080-422a-90df-59dc825f4cdb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.8899\n"
     ]
    }
   ],
   "source": [
    "acc=accuracy_score(y_val, cbtahmin)\n",
    "print(f\"Accuracy: {acc:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "id": "7af2997d-11bb-4748-800a-0a8c02ac804c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[289,   7],\n",
       "       [ 30,  10]])"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(y_val, cbtahmin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "id": "c9779a75-ca07-4fc7-87a6-9176e9efdd8c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.91      0.98      0.94       296\n",
      "         1.0       0.59      0.25      0.35        40\n",
      "\n",
      "    accuracy                           0.89       336\n",
      "   macro avg       0.75      0.61      0.65       336\n",
      "weighted avg       0.87      0.89      0.87       336\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"\\nClassification Report:\")\n",
    "print(classification_report(y_val, cbtahmin))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "id": "801bf729-ec23-4f02-9e39-11f6eeb4a294",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr, tpr, thresholds=roc_curve(y_val, y_proba_cb)\n",
    "roc_auc=auc(fpr, tpr)\n",
    "\n",
    "plt.figure(figsize=(6,5))\n",
    "plt.plot(fpr, tpr, label=f'Logistic Regression (AUC = {roc_auc:.3f})')\n",
    "plt.plot([0, 1], [0, 1], 'k--', label='Random Classifier')\n",
    "\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curve')\n",
    "plt.legend(loc='lower right')\n",
    "plt.grid(alpha=0.3)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f3ef453-29b9-47e3-bbad-e1659a2fbd25",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "id": "c5b0550c-c85e-471e-87f0-4dd3416fce3a",
   "metadata": {},
   "outputs": [],
   "source": [
    "sub=pd.read_csv('sample_submission.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "id": "093e02b2-bb8e-47f2-a172-7ae90e93fbac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>Attrition</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1677</td>\n",
       "      <td>0.119261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1678</td>\n",
       "      <td>0.119261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1679</td>\n",
       "      <td>0.119261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1680</td>\n",
       "      <td>0.119261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1681</td>\n",
       "      <td>0.119261</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     id  Attrition\n",
       "0  1677   0.119261\n",
       "1  1678   0.119261\n",
       "2  1679   0.119261\n",
       "3  1680   0.119261\n",
       "4  1681   0.119261"
      ]
     },
     "execution_count": 141,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sub.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "5decb930-99d3-4ef4-9078-1cd9f1cb5e6a",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Index: 1119 entries, 0 to 1118\n",
      "Data columns (total 49 columns):\n",
      " #   Column                             Non-Null Count  Dtype  \n",
      "---  ------                             --------------  -----  \n",
      " 0   Age                                1119 non-null   int64  \n",
      " 1   DailyRate                          1119 non-null   int64  \n",
      " 2   DistanceFromHome                   1119 non-null   int64  \n",
      " 3   Education                          1119 non-null   int64  \n",
      " 4   EnvironmentSatisfaction            1119 non-null   int64  \n",
      " 5   HourlyRate                         1119 non-null   int64  \n",
      " 6   JobInvolvement                     1119 non-null   int64  \n",
      " 7   JobLevel                           1119 non-null   int64  \n",
      " 8   JobSatisfaction                    1119 non-null   int64  \n",
      " 9   MonthlyIncome                      1119 non-null   int64  \n",
      " 10  MonthlyRate                        1119 non-null   int64  \n",
      " 11  NumCompaniesWorked                 1119 non-null   int64  \n",
      " 12  OverTime                           1119 non-null   int64  \n",
      " 13  PercentSalaryHike                  1119 non-null   int64  \n",
      " 14  PerformanceRating                  1119 non-null   int64  \n",
      " 15  RelationshipSatisfaction           1119 non-null   int64  \n",
      " 16  StockOptionLevel                   1119 non-null   int64  \n",
      " 17  TotalWorkingYears                  1119 non-null   int64  \n",
      " 18  TrainingTimesLastYear              1119 non-null   int64  \n",
      " 19  WorkLifeBalance                    1119 non-null   int64  \n",
      " 20  YearsAtCompany                     1119 non-null   int64  \n",
      " 21  YearsInCurrentRole                 1119 non-null   int64  \n",
      " 22  YearsSinceLastPromotion            1119 non-null   int64  \n",
      " 23  YearsWithCurrManager               1119 non-null   int64  \n",
      " 24  Attrition                          0 non-null      float64\n",
      " 25  time_experience                    1119 non-null   float64\n",
      " 26  no_promotion                       1119 non-null   int64  \n",
      " 27  income_experience                  1119 non-null   float64\n",
      " 28  manager_stability                  1119 non-null   float64\n",
      " 29  BusinessTravel_Travel_Frequently   1119 non-null   bool   \n",
      " 30  BusinessTravel_Travel_Rarely       1119 non-null   bool   \n",
      " 31  Department_Research & Development  1119 non-null   bool   \n",
      " 32  Department_Sales                   1119 non-null   bool   \n",
      " 33  EducationField_Life Sciences       1119 non-null   bool   \n",
      " 34  EducationField_Marketing           1119 non-null   bool   \n",
      " 35  EducationField_Medical             1119 non-null   bool   \n",
      " 36  EducationField_Other               1119 non-null   bool   \n",
      " 37  EducationField_Technical Degree    1119 non-null   bool   \n",
      " 38  Gender_Male                        1119 non-null   bool   \n",
      " 39  JobRole_Human Resources            1119 non-null   bool   \n",
      " 40  JobRole_Laboratory Technician      1119 non-null   bool   \n",
      " 41  JobRole_Manager                    1119 non-null   bool   \n",
      " 42  JobRole_Manufacturing Director     1119 non-null   bool   \n",
      " 43  JobRole_Research Director          1119 non-null   bool   \n",
      " 44  JobRole_Research Scientist         1119 non-null   bool   \n",
      " 45  JobRole_Sales Executive            1119 non-null   bool   \n",
      " 46  JobRole_Sales Representative       1119 non-null   bool   \n",
      " 47  MaritalStatus_Married              1119 non-null   bool   \n",
      " 48  MaritalStatus_Single               1119 non-null   bool   \n",
      "dtypes: bool(20), float64(4), int64(25)\n",
      "memory usage: 284.1 KB\n"
     ]
    }
   ],
   "source": [
    "test.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "73831e3e-7efc-4b3a-90a7-9f099e089a57",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_test=test.drop(['Attrition'],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "id": "75e8a7c5-7bb8-4589-befe-c1fc515235b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_proba=xg.predict_proba(x_test)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "id": "f8408e84-ced7-41f2-9eb2-8778c4d8b9fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "submission=pd.DataFrame({'id':df2['id'],'Attrition': test_proba})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "aa957e92-fb82-47f3-964c-b09b758ddeb6",
   "metadata": {},
   "outputs": [],
   "source": [
    "submission.to_csv('submission.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "id": "6cbb74e2-fd34-4140-9493-6a483d553aae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>Attrition</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1677</td>\n",
       "      <td>0.046273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1678</td>\n",
       "      <td>0.003916</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1679</td>\n",
       "      <td>0.003873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1680</td>\n",
       "      <td>0.004712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1681</td>\n",
       "      <td>0.867358</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     id  Attrition\n",
       "0  1677   0.046273\n",
       "1  1678   0.003916\n",
       "2  1679   0.003873\n",
       "3  1680   0.004712\n",
       "4  1681   0.867358"
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "submission.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "id": "959a2575-ebf2-49f4-af04-06ecba6be32c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1119, 2)\n",
      "Index(['id', 'Attrition'], dtype='object')\n",
      "id           0\n",
      "Attrition    0\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(submission.shape)\n",
    "print(submission.columns)\n",
    "print(submission.isna().sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db902384-75fc-4d54-b42f-5a7bec6d2f36",
   "metadata": {},
   "source": [
    "### Conclusion\n",
    "In this project, different machine learning models were tested to predict employee attrition.\n",
    "Logistic Regression was used as a baseline, but the XGBoost model achieved the best performance with a ROC-AUC score of about 0.82.\n",
    "\n",
    "This shows that tree-based boosting models can better capture complex patterns in employee data.\n",
    "Overall, the final model provides a reliable way to predict the probability of employee attrition."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:base] *",
   "language": "python",
   "name": "conda-base-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}